Portal de Periódicos da CAPES

Molecular Characterization of the 4′-Phosphopantothenoylcysteine Synthetase Domain of Bacterial Dfp Flavoproteins

2002; Elsevier BV; Volume: 277; Issue: 39 Linguagem: Inglês

10.1074/jbc.m206188200

1083-351X

Thomas Kupke,

RNA regulation and disease

In bacteria, coenzyme A is synthesized in five steps from pantothenate. The flavoprotein Dfp catalyzes the synthesis of the coenzyme A precursor 4′-phosphopantetheine in the presence of 4′-phosphopantothenate, cysteine, CTP, and Mg2+(Strauss, E., Kinsland, C., Ge, Y., McLafferty, F. W., and Begley, T. P. (2001) J. Biol. Chem. 276, 13513–13516). It has been shown that the NH2-terminal domain of Dfp has 4′-phosphopantothenoylcysteine decarboxylase activity (Kupke, T., Uebele, M., Schmid, D., Jung, G., Blaesse, M., and Steinbacher, S. (2000) J. Biol. Chem. 275, 31838–31846). Here I demonstrate that the COOH-terminal CoaB domain of Dfp catalyzes the synthesis of 4′-phosphopantothenoylcysteine. The exchange of conserved amino acid residues within the CoaB domain revealed that the synthesis of 4′-phosphopantothenoylcysteine occurs in two half-reactions. Using the mutant protein His-CoaB N210D the putative acyl-cytidylate intermediate of 4′-phosphopantothenate was detectable. The same intermediate was detectable for the wild-type CoaB enzyme if cysteine was omitted in the reaction mixture. Exchange of the conserved Lys289 residue, which is part of the strictly conserved289KXKK292 motif of the CoaB domain, resulted in complete loss of activity with neither the acyl-cytidylate intermediate nor 4′-phosphopantothenoylcysteine being detectable. Gel filtration experiments indicated that CoaB forms dimers. Residues that are important for dimerization are conserved in CoaB proteins from eubacteria, Archaea, and eukaryotes. In bacteria, coenzyme A is synthesized in five steps from pantothenate. The flavoprotein Dfp catalyzes the synthesis of the coenzyme A precursor 4′-phosphopantetheine in the presence of 4′-phosphopantothenate, cysteine, CTP, and Mg2+(Strauss, E., Kinsland, C., Ge, Y., McLafferty, F. W., and Begley, T. P. (2001) J. Biol. Chem. 276, 13513–13516). It has been shown that the NH2-terminal domain of Dfp has 4′-phosphopantothenoylcysteine decarboxylase activity (Kupke, T., Uebele, M., Schmid, D., Jung, G., Blaesse, M., and Steinbacher, S. (2000) J. Biol. Chem. 275, 31838–31846). Here I demonstrate that the COOH-terminal CoaB domain of Dfp catalyzes the synthesis of 4′-phosphopantothenoylcysteine. The exchange of conserved amino acid residues within the CoaB domain revealed that the synthesis of 4′-phosphopantothenoylcysteine occurs in two half-reactions. Using the mutant protein His-CoaB N210D the putative acyl-cytidylate intermediate of 4′-phosphopantothenate was detectable. The same intermediate was detectable for the wild-type CoaB enzyme if cysteine was omitted in the reaction mixture. Exchange of the conserved Lys289 residue, which is part of the strictly conserved289KXKK292 motif of the CoaB domain, resulted in complete loss of activity with neither the acyl-cytidylate intermediate nor 4′-phosphopantothenoylcysteine being detectable. Gel filtration experiments indicated that CoaB forms dimers. Residues that are important for dimerization are conserved in CoaB proteins from eubacteria, Archaea, and eukaryotes. (R)-4′-phospho-N-pantothenoylcysteine nickel-nitrilotriacetic acid, His-CoaB, MRGSHHHHHHG-Dfp Ser181–Arg406 MRGSHHHHHHGSML-CoaA electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry immobilized metal affinity chromatography wild-type N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine high performance liquid chromatography Coenzyme A is synthesized in five steps from pantothenate, and more than 40 years ago G. Brown showed that it is built from phosphorylated precursors (1Brown G.M. J. Biol. Chem. 1959; 234: 370-378Abstract Full Text PDF PubMed Google Scholar). His initial studies were mainly continued by Y. Abiko, who characterized the various enzymatic activities of coenzyme A biosynthesis in a series of papers published in 1967 and 1968 (2Abiko Y. J. Biochem. (Tokyo). 1967; 61: 300-308Crossref PubMed Scopus (29) Google Scholar, 3Abiko Y. Suzuki T. Shimizu M. J. Biochem. (Tokyo). 1967; 61: 309-312Crossref PubMed Scopus (13) Google Scholar, 4Abiko Y. J. Biochem. (Tokyo). 1967; 61: 290-299Crossref PubMed Scopus (48) Google Scholar, 5Abiko Y. Tomikawa M. Shimizu M. J. Biochem. (Tokyo). 1968; 64: 115-117Crossref PubMed Scopus (7) Google Scholar, 6Suzuki T. Abiko Y. Shimizu M. J Biochem. (Tokyo). 1967; 62: 642-649Crossref PubMed Scopus (18) Google Scholar). The conclusion of both pioneer works was that coenzyme A is synthesized in bacterial and in mammalian systems as follows (see Fig. 1). In the first step, pantothenate is phosphorylated to 4′-phosphopantothenate, which is then converted by addition of cysteine to (R)-4′-phospho-N-pantothenoylcysteine (PPC,1 “coupling reaction”). PPC is then decarboxylated to 4′-phosphopantetheine, which is converted to coenzyme A in two further steps. Although the pathway for coenzyme A biosynthesis has been well established since the work of Brown and Abiko, purification and detailed characterization of the enzymes involved turned out to be very difficult. Since the coenzyme A biosynthetic genes are essential, their identification and cloning was also a great challenge. Starting with cloning of the pantothenate kinase gene (coaA) fromEscherichia coli in 1992 (7Song W.J. Jackowski S. J. Bacteriol. 1992; 174: 6411-6417Crossref PubMed Google Scholar), it took 10 years to identify all the genes of the coenzyme A biosynthetic pathway in E. coli (8Kupke T. Uebele M. Schmid D. Jung G. Blaesse M. Steinbacher S. J. Biol. Chem. 2000; 275: 31838-31846Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar, 9Strauss E. Kinsland C., Ge, Y. McLafferty F.W. Begley T.P. J. Biol. Chem. 2001; 276: 13513-13516Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar, 10Mishra P. Park P.K. Drueckhammer D.G. J. Bacteriol. 2001; 183: 2774-2778Crossref PubMed Scopus (65) Google Scholar, 11Geerlof A. Lewendon A. Shaw W.V. J. Biol. Chem. 1999; 274: 27105-27111Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar) and to identify the human genes by comparative genetics (12Daugherty M. Polanuyer B. Farrell M. Scholle M. Lykidis A., De Crécy-Lagard V. Osterman A. J. Biol. Chem. 2002; 277: 21431-21439Abstract Full Text Full Text PDF PubMed Scopus (189) Google Scholar). In bacteria, the two-step conversion of 4′-phosphopantothenate to 4′-phosphopantetheine (peptide bond formation between 4′-phosphopantothenate and cysteine and then decarboxylation of the formed PPC) is catalyzed by the Dfp flavoproteins (Fig.1 and Refs. 8Kupke T. Uebele M. Schmid D. Jung G. Blaesse M. Steinbacher S. J. Biol. Chem. 2000; 275: 31838-31846Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar and 9Strauss E. Kinsland C., Ge, Y. McLafferty F.W. Begley T.P. J. Biol. Chem. 2001; 276: 13513-13516Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar), which were first described by Spitzer et al. (13Spitzer E.D. Weiss B. J. Bacteriol. 1985; 164: 994-1003Crossref PubMed Google Scholar, 14Spitzer E.D. Jimenez-Billini H.E. Weiss B. J. Bacteriol. 1988; 170: 872-876Crossref PubMed Google Scholar). Molecular characterization showed that the decarboxylase activity resides in the NH2-terminal CoaC domain of Dfp, indicating that the COOH-terminal (CoaB) domain is involved in synthesis of PPC (8Kupke T. Uebele M. Schmid D. Jung G. Blaesse M. Steinbacher S. J. Biol. Chem. 2000; 275: 31838-31846Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). PPC decarboxylases bind the cofactor FMN, and during the last 2 years it has been shown that decarboxylation of PPC follows the mechanism “decarboxylation by initial oxidation” of the so-called LanD enzymes (8Kupke T. Uebele M. Schmid D. Jung G. Blaesse M. Steinbacher S. J. Biol. Chem. 2000; 275: 31838-31846Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar, 15Blaesse M. Kupke T. Huber R. Steinbacher S. EMBO J. 2000; 19: 6299-6310Crossref PubMed Scopus (83) Google Scholar, 16Hernández-Acosta P. Schmid D.G. Jung G. Culiáñez-Macià F.A. Kupke T. J. Biol. Chem. 2002; 277: 20490-20498Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar, 17Kupke T. Kempter C. Gnau V. Jung G. Götz F. J. Biol. Chem. 1994; 269: 5653-5659Abstract Full Text PDF PubMed Google Scholar, 18Strauss E. Begley T.P. J. Am. Chem. Soc. 2001; 123: 6449-6450Crossref PubMed Scopus (36) Google Scholar). In 2002 it was confirmed that the human PPC decarboxylase is also a flavoprotein (12Daugherty M. Polanuyer B. Farrell M. Scholle M. Lykidis A., De Crécy-Lagard V. Osterman A. J. Biol. Chem. 2002; 277: 21431-21439Abstract Full Text Full Text PDF PubMed Scopus (189) Google Scholar). Earlier results classified bacterial and mammalian PPC decarboxylases as pyruvoyl-dependent enzymes (19Scandurra R. Barboni E. Granata F. Pensa B. Costa M. Eur. J. Biochem. 1974; 49: 1-9Crossref PubMed Scopus (22) Google Scholar, 20Scandurra R. Politi L. Santoro L. Consalvi V. FEBS Lett. 1987; 212: 79-82Crossref PubMed Scopus (10) Google Scholar, 21Yang H. Abeles R.H. Biochemistry. 1987; 26: 4076-4081Crossref PubMed Scopus (22) Google Scholar), although there is no free amino group in the substrate. Begley et al. (22Begley T.P. Kinsland C. Strauss E. Vitam. Horm. 2001; 61: 157-171Crossref PubMed Google Scholar) first proposed a new mechanism for a pyruvoyl-dependent decarboxylation involving a N-S acyl shift to unmask the amino group, but later they confirmed that the flavoprotein Dfp catalyzes the decarboxylation of PPC as had been shown earlier by Kupke et al. (8Kupke T. Uebele M. Schmid D. Jung G. Blaesse M. Steinbacher S. J. Biol. Chem. 2000; 275: 31838-31846Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar, 9Strauss E. Kinsland C., Ge, Y. McLafferty F.W. Begley T.P. J. Biol. Chem. 2001; 276: 13513-13516Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar). Until now the synthesis of PPC by Dfp has only been shown indirectly by measurement of released14CO2 froml-1-[14C]cysteine in the presence of 4′-phosphopantothenate, CTP, and Mg2+ (9Strauss E. Kinsland C., Ge, Y. McLafferty F.W. Begley T.P. J. Biol. Chem. 2001; 276: 13513-13516Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar). BacterialPPC synthesis requires the cofactor CTP (1Brown G.M. J. Biol. Chem. 1959; 234: 370-378Abstract Full Text PDF PubMed Google Scholar), which is converted during the reaction to CMP and inorganic pyrophosphate indicating the formation of an activated acyl-cytidylate as intermediate (Fig. 1 and Ref. 9Strauss E. Kinsland C., Ge, Y. McLafferty F.W. Begley T.P. J. Biol. Chem. 2001; 276: 13513-13516Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar). HumanPPC synthetase uses ATP for the coupling reaction more efficiently than CTP. The synthesis of PPC was studied in this case by the release of inorganic pyrophosphate indicating that the activation mechanism of the human enzyme is comparable to that of bacterial enzymes (12Daugherty M. Polanuyer B. Farrell M. Scholle M. Lykidis A., De Crécy-Lagard V. Osterman A. J. Biol. Chem. 2002; 277: 21431-21439Abstract Full Text Full Text PDF PubMed Scopus (189) Google Scholar). However, Abiko et al. (5Abiko Y. Tomikawa M. Shimizu M. J. Biochem. (Tokyo). 1968; 64: 115-117Crossref PubMed Scopus (7) Google Scholar) showed that the PPC synthetase from rat liver converts ATP to ADP and phosphate (and not to AMP and pyrophosphate). In this study, a molecular characterization of the COOH-terminal CoaC domain of Dfp was undertaken, and its ability to catalyze the synthesis of PPC is demonstrated. In contrast to previous studies onPPC synthetases (9Strauss E. Kinsland C., Ge, Y. McLafferty F.W. Begley T.P. J. Biol. Chem. 2001; 276: 13513-13516Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar, 12Daugherty M. Polanuyer B. Farrell M. Scholle M. Lykidis A., De Crécy-Lagard V. Osterman A. J. Biol. Chem. 2002; 277: 21431-21439Abstract Full Text Full Text PDF PubMed Scopus (189) Google Scholar), the formation of PPC is directly shown. Conserved sequence motifs in the CoaC domain are investigated by site-directed mutagenesis, continuing the molecular characterization of the Dfp proteins (8Kupke T. Uebele M. Schmid D. Jung G. Blaesse M. Steinbacher S. J. Biol. Chem. 2000; 275: 31838-31846Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar, 23Kupke T. J. Biol. Chem. 2001; 276: 27597-27604Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). These studies led to detection of the proposed activated acyl-cytidylate intermediate, 4′-phosphopantothenoyl-CMP, and show that the synthesis ofPPC occurs in two half-reactions. Furthermore, it is shown that CoaB forms homodimers and that residues responsible for dimerization are conserved in CoaB (Dfp) proteins from all kingdoms of life. PCR amplifications were performed with Vent-DNA polymerase (New England Biolabs). The entire sequences of the coaA and coaB coding regions of the constructed plasmids were verified. Oligonucleotides were purchased from MWG Biotech. The 3′-part of thedfp gene encoding the COOH-terminal CoaB domain Ser181–Arg406 of Dfp was amplified by PCR and cloned into the single BamHI site of the expression vector pQE8 (Qiagen). For PCR amplification, the template pQE12 dfp(8Kupke T. Uebele M. Schmid D. Jung G. Blaesse M. Steinbacher S. J. Biol. Chem. 2000; 275: 31838-31846Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar) and the primers (i) forward, 5′-GCGGTAGCGCATAGATCTCCCGTCAACGACC-3′ (the introduced BglII site is underlined) and (ii) reverse, 5′-GGTCATTACTGGATCTATCAACAGG-3′ were used. The reverse primer binds downstream of the BglII site of pQE12dfp. The amplified coaB gene was digested withBglII and cloned into pQE8 BamHI. The pQE8-derived plasmids were transformed into the expression strainE. coli M15 (pREP4) (Qiagen) by electroporation. The expression plasmid pQE8 coaB encodes an NH2-terminal His tag fusion protein of the CoaB protein (His-CoaB: MRGSHHHHHHG-Dfp Ser181–Arg406). All point mutations were first introduced into pQE12 dfp by using sequential PCR and appropriate mutagenesis primers as recently described (23Kupke T. J. Biol. Chem. 2001; 276: 27597-27604Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar, 24Cormack B. Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Seidman J.G. Smith J.A. Struhl K. Current Protocols in Molecular Biology. 1. John Wiley & Sons, Inc., New York1991: 8.5.1-8.5.9Google Scholar). Using the constructed mutant pQE12 dfpplasmids as templates, the mutant coaB genes were amplified and then cloned into pQE8 BamHI as described above for wtcoaB. Chromosomal DNA fromE. coli TB1 (New England Biolabs) was purified using the Qiagen Blood & Cell Culture DNA Mini kit. For cloning, thecoaA gene (7Song W.J. Jackowski S. J. Bacteriol. 1992; 174: 6411-6417Crossref PubMed Google Scholar) was amplified by PCR using the oligonucleotides (i) forward, 5′-GCTATGACCGCCGGATCCATGCTTATGAGTA-3′ and (ii) reverse, 5′-GAAAGGGGAGTATTGGATCCCCTGCAAATT-3′ as primers (introducedBamHI sites are underlined) and the purified chromosomal DNA as template. The amplified coaA gene was cloned into pQE8BamHI and then transformed in E. coli M15 (pREP4) as described above. The expression plasmid pQE8 coaA encodes an NH2-terminal His tag fusion protein of the CoaA protein (His-CoaA: MRGSHHHHHHGSML-CoaA). E. coli M15 (pREP4, pQE8) cells were grown in the presence of 100 μg/ml ampicillin and 25 μg/ml kanamycin in 0.5 liters of B-broth (10 g of casein hydrolysate 140 (Invitrogen), 5 g of yeast extract (Difco), 5 g of NaCl, 1 g of glucose, and 1 g of K2HPO4/liter, pH 7.3) in 2-liter shaker flasks. At A 578 = 0.4, the cells were induced with 1 mm isopropyl-β-d-thiogalactopyranoside and harvested 2 h after induction. The growth temperature was 37 °C. For purification of His-CoaA and His-CoaB proteins, 500 ml of isopropyl-β-d-thiogalactopyranoside-inducedE. coli M15 (pREP4, pQE8 coaA/coaB) cells were harvested and disrupted by sonication in 10 ml of 20 mmTris-HCl (pH 8.0). 0.65–1.3 ml of the cleared lysates obtained by two centrifugation steps (each 20 min at 30,000 × g at 4 °C) were applied to Ni-NTA spin columns (Qiagen) equilibrated with column buffer (20 mm Tris-HCl, pH 8.0, 10 mmimidazole, 300 mm NaCl). The spin columns were then washed twice with 0.65 ml of column buffer. His-CoaA, His-CoaB, and mutant His-CoaB proteins were eluted with 0.16 ml of column buffer containing 250 mm instead of 10 mm imidazole. The Ni-NTA spin columns were centrifuged at room temperature at only 240 ×g to enable effective binding of the His tag proteins. For gel filtration, a 25-μl aliquot of the Ni-NTA eluate was subjected to a Superdex 200 PC 3.2/30 column equilibrated in running buffer (20 mm Tris-HCl, pH 8.0, 200 mm NaCl) at a flow rate of 40 μl/min. The Superdex 200 PC 3.2/30 column and the standard proteins used for calibration were obtained from Amersham Biosciences. For activity assays (see below), the Ni-NTA eluates were used. Since 4′-phosphopantothenate is not commercially available, it was synthesized enzymatically by adding His-CoaA, pantothenate, and ATP to the His-CoaB assay mixtures. Therefore, 0.8 ml of CoaB assay mixtures contained 5 mm pantothenate, 2.5 mm MgCl2, 5 mm ATP, 5 mm CTP, 10 mm cysteine hydrochloride, 10 mm dithiothreitol, 50 mm Tris, pH 8.0, His-CoaA (∼10–15 μg), and either wt His-CoaB or mutant His-CoaB proteins in the range of 3–20 μg. After 45 min of incubation at 37 °C, the reaction mixtures were kept at −80 °C and then were successively separated by reversed phase chromatography with a μRPC C2/C18 SC 2.1/10 column on a SMART system (Amersham Biosciences). Compounds were eluted with a linear gradient of 0–50% acetonitrile, 0.1% trifluoroacetic acid in 5.8 ml with a flow rate of 200 μl/min. The absorbance was measured simultaneously at 214, 260, and 280 nm to enable identification of acyl-cytidylate intermediates. Proteins were separated using Tricine-sodium dodecyl sulfate-polyacrylamide (10%) gel electrophoresis under reducing conditions (25Schägger H. Jagow G. Anal. Biochem. 1987; 166: 368-379Crossref PubMed Scopus (10503) Google Scholar). Prestained protein molecular weight standards were obtained from New England Biolabs. Recently the alignment of the FMN binding CoaC domains of sequenced bacterial Dfp proteins was presented, and a PPC decarboxylase signature was defined (23Kupke T. J. Biol. Chem. 2001; 276: 27597-27604Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). Here the sequence comparison is extended to the COOH-terminal domain of Dfp proteins (Fig.2 A). There are a lot of conserved sequence motifs within the CoaB domains with the210NXSSGK215 and the289KXKK292 motifs being the most noteworthy. In eukaryotes (and a few bacteria), the PPC decarboxylase and synthetase activities are not fused (12Daugherty M. Polanuyer B. Farrell M. Scholle M. Lykidis A., De Crécy-Lagard V. Osterman A. J. Biol. Chem. 2002; 277: 21431-21439Abstract Full Text Full Text PDF PubMed Scopus (189) Google Scholar, 26Kupke T. Hernández-Acosta P. Steinbacher S. Culiáñez-Macià F.A. J. Biol. Chem. 2001; 276: 19190-19196Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). Human and plant PPC decarboxylases share the bacterialPPC decarboxylase signature, whereas there is only marginal similarity of the eukaryotic PPC synthetases with the bacterial CoaB domains (12Daugherty M. Polanuyer B. Farrell M. Scholle M. Lykidis A., De Crécy-Lagard V. Osterman A. J. Biol. Chem. 2002; 277: 21431-21439Abstract Full Text Full Text PDF PubMed Scopus (189) Google Scholar, 26Kupke T. Hernández-Acosta P. Steinbacher S. Culiáñez-Macià F.A. J. Biol. Chem. 2001; 276: 19190-19196Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). The sequence comparison shown is the basis for the site-directed mutagenesis studies of the CoaB domain presented below. Interestingly, it turned out that despite the low similarity of the PPC synthetase domains, residues that are proposed to be important for dimer formation of CoaB are conserved in CoaB proteins from eukaryotes and Archaea (Fig. 2 B and see below). It appears that the first Lys residue of the KXKK motif is also conserved in CoaB from eubacteria, eukaryotes, and Archaea (Fig. 2 B). However, the sequence similarity is very low, and biochemical studies are required to prove that the Lys residue has the same function in all CoaB proteins. CoaB and mutant coaB genes were expressed as His tag fusion proteins, purified from the corresponding E. coli clones by immobilized metal affinity chromatography (IMAC) and additionally purified by gel filtration (Figs. 3 and4). Gel filtration was also used to determine the apparent molecular weight and to elucidate whether native His-CoaB forms monomers or homomultimers. A comparison with standard proteins revealed an apparent molecular mass of 42 kDa for native purified His-CoaB, indicating that His-CoaB forms homodimers (Fig. 4, molecular mass of His-CoaB is 26.1 kDa). Some of the analyzed mutant His-CoaB proteins (T194V, T198V, D203N, and A275V) showed a significantly increased elution volume that corresponds exactly to the molecular weight of monomeric His-CoaB. Also the elution volume of His-CoaB N210D is increased, but it appears that the mass of this protein is slightly increased compared with monomeric His-CoaB. In general, molecular weight determination of proteins by gel filtration is not precise, but the differences of the elution volumes between wt His-CoaB and mutant His-CoaB proteins clearly show that wt enzyme forms homodimers, while several of the mutants are monomeric. The residues that are important for dimerization of E. coliCoaB are conserved in eukaryotic PPC synthetases (Fig.2 B). It appears that residues Thr194–Asn210 of E. coli CoaB form a dimerization motif.Figure 4Gel filtration of mutant His-CoaB proteins. IMAC-purified His-CoaB proteins were separated by gel filtration on a Superdex 200 PC 3.2/30 column, and the elution was monitored by absorbance at 278 nm. The mutant proteins were analyzed under comparable conditions together with wild-type His-CoaB in two sets of experiments (A, His-CoaB wt, His-CoaB N210D, His-CoaB S212A, His-CoaB K215Q, His-CoaB D279N, His-CoaB D279E, His-CoaB K289Q, His-CoaB K291Q, and His-CoaB K292Q; B, His-CoaB wt, His-CoaB T194V, His-CoaB T198V, His-CoaB D203N, His-CoaB R206Q, His-CoaB A275S, His-CoaB A275V, His-CoaB A276V, His-CoaB D309N, and His-CoaB F327L). The determined elution volume of wild-type His-CoaB corresponds to a molecular mass of 42 kDa. The mutations T194V, T198V, D203N, N210D, and A275V (framed) led to significant changes of the elution volume. The gel filtration column used was calibrated with standard proteins to correlate the elution volume with molecular weight information as described recently (8Kupke T. Uebele M. Schmid D. Jung G. Blaesse M. Steinbacher S. J. Biol. Chem. 2000; 275: 31838-31846Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar).View Large Image Figure ViewerDownload Hi-res image Download (PPT) Recently we showed that Dfp is a homododecameric member of the homooligomeric flavin-containing Cys decarboxylase protein family (8Kupke T. Uebele M. Schmid D. Jung G. Blaesse M. Steinbacher S. J. Biol. Chem. 2000; 275: 31838-31846Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar,15Blaesse M. Kupke T. Huber R. Steinbacher S. EMBO J. 2000; 19: 6299-6310Crossref PubMed Scopus (83) Google Scholar) and that the NH2-terminal CoaC domain is responsible for homododecamer formation (8Kupke T. Uebele M. Schmid D. Jung G. Blaesse M. Steinbacher S. J. Biol. Chem. 2000; 275: 31838-31846Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar, 23Kupke T. J. Biol. Chem. 2001; 276: 27597-27604Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). Mutations that prevent dimerization of CoaB do not prevent the formation of homododecameric Dfp (data not shown). Including the new results, I propose that within the Dfp homododecamers two PPC synthetase domains contact each other. Strausset al. (9Strauss E. Kinsland C., Ge, Y. McLafferty F.W. Begley T.P. J. Biol. Chem. 2001; 276: 13513-13516Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar) have shown that Dfp proteins are able to convert 4′-phosphopantothenate to 4′-phosphopantetheine when cysteine, dithiothreitol, CTP, and Mg2+ are present in the assay mixture, and they verified that Dfp catalyzes the decarboxylation ofPPC to 4′-phosphopantetheine. Considering the known biosynthetic pathway for coenzyme A, these results show that Dfp is a bifunctional enzyme and catalyzes the synthesis of PPC and subsequently the decarboxylation of PPC to 4′-phosphopantetheine. However, Strauss et al. (9Strauss E. Kinsland C., Ge, Y. McLafferty F.W. Begley T.P. J. Biol. Chem. 2001; 276: 13513-13516Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar) could not directly detect PPC as the intermediate of the Dfp reaction. To analyze the biochemistry of the PPC synthetase activity of E. coli, it is necessary to separate PPC synthetase and PPC decarboxylase activities of Dfp. This can be achieved by disrupting the PPC decarboxylase activity or by separating the bifunctional Dfp protein into the two proposed domains, CoaB and CoaC. Both approaches were tried in the present study and were successful. Using the published HPLC method for the detection of PPC (8Kupke T. Uebele M. Schmid D. Jung G. Blaesse M. Steinbacher S. J. Biol. Chem. 2000; 275: 31838-31846Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar), I was able to show that the COOH-terminal CoaB domain of Dfp synthesizes PPC from 4′-phosphopantothenate and cysteine (Figs. Figure 5, Figure 6, Figure 7). Biosynthesis of PPC was not only verified by comparison of the determined retention volume with that of synthetic PPC but also by electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry (ESI-FTICR-MS) as described recently (8Kupke T. Uebele M. Schmid D. Jung G. Blaesse M. Steinbacher S. J. Biol. Chem. 2000; 275: 31838-31846Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). Them/z value for PPC synthesized by CoaB was determined to be 403.09, which is in accordance with the theoretical mass of [PPC + H]+ = 403.0935 Da. Synthesis of PPC was also observed when Dfp C158S (26Kupke T. Hernández-Acosta P. Steinbacher S. Culiáñez-Macià F.A. J. Biol. Chem. 2001; 276: 19190-19196Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar) was used instead of CoaB (Fig. 6). Recently it has been shown that the C158S mutation (Cys158 is within the amino-terminal CoaC domain of Dfp) inhibits the PPC decarboxylase activity (23Kupke T. J. Biol. Chem. 2001; 276: 27597-27604Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar).Figure 6PPC synthetase activity of Dfp C158S and mutant CoaB proteins. Biosynthesis of PPC was analyzed using the described HPLC-based assay. PPC was detectable when wt His-CoaB, His-CoaB S212A, His-CoaB K215Q, His-CoaB K291Q, or His-CoaB K292Q was used. However, no PPC was detectable when the mutant proteins His-CoaB N210D or His-CoaB K289Q were used for the assays. Interestingly, incubation with His-CoaB N210D led to a compound (labeled with an asterisk) with high absorbances at 260 and 280 nm (not shown). This compound was later identified to be the intermediate of the CoaB reaction (compare Fig.7). Synthesis of PPC could also be achieved with the mutant Dfp C158S protein, which is not able to decarboxylate PPC to 4′-phosphopantetheine. Synthetic PPC was used in the control experiment. The retention time of PPC was shifted to lower values compared with the data presented in Fig. 5. PPC already eluted during washing the column with H2O, 0.1% trifluoroacetic acid (gradient started at 17 min). This is due to the increased number of applications of the column used and the large amounts of compounds applied to the column during the assays.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 5PPC synthetase activity of the CoaB domain. The synthesis of 4′-phosphopantothenoylcysteine from 4′-phosphopantothenate and cysteine was analyzed using an HPLC-based assay. 4′-Phosphopantothenate was synthesized in situ by incubation of pantothenate with ATP, Mg2+, and pantothenate kinase (His-CoaA). A, PPC is synthesized in the presence of pantothenic acid, Mg2+, ATP, CTP, cysteine, dithiothreitol, His-CoaA, and wt His-CoaB. B, when His-CoaB is omitted from the reaction mixture, no PPC is detectable. IMAC purification of the pantothenate kinase His-CoaA used was verified by SDS-PAGE as shown in the inset of B. C, no pantothenoylcysteine was detectable if pantothenic acid was incubated in the presence of CTP, cysteine, dithiothreitol, and Mg2+ with His-CoaB, indicating that only 4′-phosphopantothenate, and not pantothenate, is a substrate of CoaB.D, synthetic PPC was used as standard substance to evaluate the HPLC assays.View Large Image Figure ViewerDownload Hi-res image Download (PPT) In this study, the motifs 210NXSSGK215 and289KXKK292 of CoaB were investigated in more detail by site-directed mutagenesis. The mutant proteins were purified and characterized by gel filtration (see above), and thePPC synthetase activity was analyzed (Fig. 6). Although the assay used is not suitable for the determination of kinetic parameters, it can be applied to identify amino acid residues that are crucial for activity. Introduction of the point mutations N210D and K289Q led to complete loss of PPC synthetase activity. However, His-CoaB N210D was able to form 4′-phosphopantothenoyl-CMP, the proposed intermediate of PPC biosynthesis (Figs. 6 and7 and see below). The conclusion is thatPPC biosynthesis occurs in two half-reactions and that the residue Lys289 is important for formation of the activated acyl-cytidylate intermediate (Fig. 1 B). Furthermore, it is reasonable to suggest that Lys289 may either be involved in binding the phosphate group of 4′-phosphopantothenate or in binding the phosphate groups of the cofactor CTP. The activity of the mutant His-CoaB N210D protein was analyzed in more detail (Fig. 7). Only very small amounts of PPC were synthesized by His-CoaB N210D. However, an intermediate with an increased retention time compared with PPC could be identified. This intermediate was only present in very small amounts, and repeated attempts to determine the molecular mass of this compound failed. Therefore, I tried to determine whether 4′-phosphopantothenate or cysteine is converted to this intermediate. The results obtained clearly show that phosphopantothenate, but not cysteine, is converted to the intermediate (Fig. 7). As expected, this intermediate could also be detected with wt His-CoaB when cysteine was omitted in the assay mixture (Fig. 7). Further characterization of the intermediate was possible by following the elution of the compound from the used C2/C18 column at various wavelengths. This revealed high absorbances of the compound at 260 nm (not shown) and 280 nm (Fig. 7). Pantothenate does not absorb in this UV range, and the determined ratio of the absorbances,A 280/A 260 = 2.0, showed that the intermediate is derived from CTP and not ATP. Taking all these data together and including the observation of Strauss et al. (9Strauss E. Kinsland C., Ge, Y. McLafferty F.W. Begley T.P. J. Biol. Chem. 2001; 276: 13513-13516Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar) that CTP is converted to CMP and inorganic pyrophosphate during the Dfp reaction, I conclude that the detected compound is the proposed 4′-phosphopantothenoyl-CMP intermediate (Fig. 1 B). If this intermediate is not attacked by cysteine, it will be present in an enzyme-bound form and will be released from the enzyme under acidic conditions as has been observed in the experiments shown. The conclusion is that the mutation N210D either impairs binding of cysteine or influences the nucleophilic attack on the intermediate by cysteine (Fig. 1). The experiments also indicate that formation of 4′-phosphopantothenoyl-CMP does not require a dimeric structure of CoaB since the N210D mutant is a monomer. The increased mass of purified His-CoaB N210D compared with the determined mass of monomeric His-CoaB may be explained by binding of the acyl-cytidylate. Pantothenate synthetase catalyzes the condensation of pantoate with β-alanine in an ATP-dependent way in two half-reactions via the pantoyl-adenylate intermediate (27Miyatake K. Nakano Y. Kitaoka S. Methods Enzymol. 1979; 62: 215-219Crossref PubMed Scopus (25) Google Scholar, 28Zheng R. Blanchard J.S. Biochemistry. 2001; 40: 12904-12912Crossref PubMed Scopus (93) Google Scholar). β-Alanine is derived from aspartate by decarboxylation catalyzed by the pyruvoyl-dependent aspartate 1-decarboxylase (29van Poelje P.D. Snell E.E. Annu. Rev. Biochem. 1990; 59: 29-59Crossref PubMed Scopus (203) Google Scholar, 30Williamson J.M. Brown G.M. J. Biol. Chem. 1979; 254: 8074-8082Abstract Full Text PDF PubMed Google Scholar). It is interesting to compare this pathway with the biosynthesis of 4′-phosphopantetheine catalyzed by the Dfp protein. Both synthetases, pantothenate synthetase and CoaB, are dimers. The catalysis of peptide bond formation is similar and depends on the formation of activated acyl-cytidylate or acyl-adenylate intermediates. However, the involved decarboxylation reactions are different. Aspartate is decarboxylated before formation of the peptide bond, whereas decarboxylation of cysteine is not observed. The cysteamine residue of 4′-phosphopantetheine is introduced by FMN-dependent decarboxylation of PPC by the NH2-terminal CoaC domain of Dfp. Recently it was shown that the structure of the NH2-terminal domain of pantothenate synthetase is very similar to class I aminoacyl-tRNA synthetases and that pantothenate synthetase belongs to the cytidylyltransferase superfamily (31von Delft F. Lewendon A. Dhanaraj V. Blundell T.L. Abell C. Smith A.G. Structure (Camb.). 2001; 9: 439-450Abstract Full Text Full Text PDF PubMed Scopus (71) Google Scholar). It will be interesting to learn more about the three-dimensional structures of Dfp and CoaB by x-ray diffraction methods to elucidate how CTP is bound by the enzyme and to gain more insights into the reaction mechanism. Dfp is a bifunctional enzyme catalyzing the synthesis of 4′-phosphopantetheine in a multistep process from 4′-phosphopantothenate and cysteine. In the first step, 4′-phosphopantothenate is activated by reaction with CTP. The 4′-phosphopantothenoyl-cytidylate formed is attacked by cysteine, andPPC is synthesized. These reactions occur in the COOH-terminal CoaB domain of Dfp. The next step is the FMN-dependent oxidative decarboxylation of PPC to 4′-phosphopantothenoylaminoethenethiol, which is then reduced to 4′-phosphopantetheine (16Hernández-Acosta P. Schmid D.G. Jung G. Culiáñez-Macià F.A. Kupke T. J. Biol. Chem. 2002; 277: 20490-20498Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar). Oxidative decarboxylation of peptidylcysteines has been detected as an important step in the biosynthesis of the lantibiotic epidermin (17Kupke T. Kempter C. Gnau V. Jung G. Götz F. J. Biol. Chem. 1994; 269: 5653-5659Abstract Full Text PDF PubMed Google Scholar). The decarboxylation reaction is catalyzed by the NH2-terminal CoaC domain, and presently it is not known how CoaB and CoaC domains interact to synthesize 4′-phosphopantetheine most effectively.