Cloning, Sequencing, Heterologous Expression, Purification, and Characterization of Adenosylcobalamin-dependentd-Lysine 5,6-Aminomutase from Clostridium sticklandii
2000; Elsevier BV; Volume: 275; Issue: 1 Linguagem: Inglês
10.1074/jbc.275.1.106
ISSN1083-351X
AutoresChristopher H. Chang, Perry A. Frey,
Tópico(s)Heme Oxygenase-1 and Carbon Monoxide
Resumod-Lysine 5,6-aminomutase fromClostridium sticklandii catalyzes the 1,2-shift of the ε-amino group of d-lysine and reverse migration of C5(H). The two genes encoding 5,6-aminomutase have been cloned, sequenced, and expressed in Escherchia coli. They are adjacent on theClostridial chromosome and encode polypeptides of 57.3 and 29.2 kilodaltons. The predicted amino acid sequence includes a conserved base-off 5′-deoxyadenosylcobalamin binding motif and a 3-cysteine cluster in the small subunit, as well as a P-loop sequence in the large subunit. Activity of the recombinant enzyme exceeds that of the 5,6-aminomutase purified from C. sticklandii by 6-fold, presumably due to the absence of bound, inactive corrinoids in the recombinant enzyme. The K m values for adenosylcobalamin and pyridoxal 5′-phosphate are 6.6 and 1.0 μm, respectively. ATP does not have a regulatory effect on the recombinant protein. The rapid turnover associated inactivation reported for the enzyme purified from Clostridium is also seen with the recombinant form. Aminomutase activity does not depend on structural or catalytic metal ions. Electron paramagnetic resonance experiments with [15N-dimethylbenz-imidazole]adenosylcobalamin demonstrate base-off binding, consistent with other B12-dependent enzymes that break unactivated C—H bonds. d-Lysine 5,6-aminomutase fromClostridium sticklandii catalyzes the 1,2-shift of the ε-amino group of d-lysine and reverse migration of C5(H). The two genes encoding 5,6-aminomutase have been cloned, sequenced, and expressed in Escherchia coli. They are adjacent on theClostridial chromosome and encode polypeptides of 57.3 and 29.2 kilodaltons. The predicted amino acid sequence includes a conserved base-off 5′-deoxyadenosylcobalamin binding motif and a 3-cysteine cluster in the small subunit, as well as a P-loop sequence in the large subunit. Activity of the recombinant enzyme exceeds that of the 5,6-aminomutase purified from C. sticklandii by 6-fold, presumably due to the absence of bound, inactive corrinoids in the recombinant enzyme. The K m values for adenosylcobalamin and pyridoxal 5′-phosphate are 6.6 and 1.0 μm, respectively. ATP does not have a regulatory effect on the recombinant protein. The rapid turnover associated inactivation reported for the enzyme purified from Clostridium is also seen with the recombinant form. Aminomutase activity does not depend on structural or catalytic metal ions. Electron paramagnetic resonance experiments with [15N-dimethylbenz-imidazole]adenosylcobalamin demonstrate base-off binding, consistent with other B12-dependent enzymes that break unactivated C—H bonds. pyridoxal 5′-phosphate 5′-deoxyadenosylcobalamin 5,6-dimethylbenzimidazole polyacrylamide gel electrophoresis 4-(2-hydroxyethyl)-1-piperazinepropanesulfonic acid polymerase chain reaction kilobase pair(s) base pair(s) β-mercaptoethanol Clostridium sticklandii ferments lysine to acetic acid, butyric acid, and ammonia. There are two separate pathways of lysine catabolism in this bacterium, which differ for l- andd-lysine (1Stadtman T.C. Adv. Enzymol. 1973; 38: 413-447Google Scholar). The first committed step ind-lysine catabolism is catalyzed by d-lysine 5,6-aminomutase. The enzyme can be purified as a 170-kDa complex of 55- and 30-kDa subunits. The reaction catalyzed is a 1,2-migration of thed-lysine ε-amino group to the δ-carbon with concomitant reverse migration of a hydrogen atom to produce 2,5-diaminohexanoic acid (2Morley C.G.D. Stadtman T.C. Biochemistry. 1971; 10: 2325-2329Crossref PubMed Scopus (21) Google Scholar). The overall reaction requires cleavage of unactivated C—H and C—N bonds, both of which are strong and present an energetic challenge. d-Lysine 5,6-aminomutase activity requires both adenosylcobalamin and pyridoxal 5′-phosphate (3Morley C.G.D. Stadtman T.C. Biochemistry. 1970; 9: 4890-4900Crossref PubMed Scopus (43) Google Scholar, 4Morley C.G.D. Stadtman T.C. Biochemistry. 1972; 11: 600-605Crossref PubMed Scopus (17) Google Scholar). The enzyme also catalyzes a similar amino group migration with l-β-lysine as the substrate (5Baker J.J. van der Drift C. Stadtman T.C. Biochemistry. 1973; 12: 1054-1063Crossref PubMed Scopus (42) Google Scholar, 6Baker J.J. Stadtman T.C. Dolphin D. B12. 2. John Wiley & Sons, Inc., New York1984: 203-231Google Scholar). Early preparations from C. sticklandii included an ∼80 kDa auxiliary sulfhydryl activating protein (E2) in complex with the tetrameric aminomutase (E 1). E2 exhibited adenosylcobalamin synthetase activity when isolated separately, and it was found to activate and transfer radioactivity to E1 when [8-14C]ATP was included in aminomutase reaction mixtures (5Baker J.J. van der Drift C. Stadtman T.C. Biochemistry. 1973; 12: 1054-1063Crossref PubMed Scopus (42) Google Scholar). ATP played an additional role of an allosteric activator when E2 was present in assay mixtures. Tritium was transferred from 5′-[3H2]adenosylcobalamin to substrate and product, as well as from substrate and product to the cofactor (2Morley C.G.D. Stadtman T.C. Biochemistry. 1971; 10: 2325-2329Crossref PubMed Scopus (21) Google Scholar). The overall reaction appears similar to that catalyzed by PLP1-dependentl-lysine 2,3-aminomutase from Clostridium subterminale (7Chirpich T.P. Zappia V. Costilow R.N. Barker H.A. J. Biol. Chem. 1970; 245: 1778-1789Abstract Full Text PDF PubMed Google Scholar, 8Baraniak J. Moss M.L. Frey P.A. J. Biol. Chem. 1989; 264: 1357-1360Abstract Full Text PDF PubMed Google Scholar, 9Petrovich R.M. Ruzicka F.J. Reed G.H. Frey P.A. Biochemistry. 1992; 31: 10774-10781Crossref PubMed Scopus (78) Google Scholar); however, the [4Fe-4S] cluster andS-adenosyl-l-methionine of thel-lysine 2,3-aminomutase appear to be replaced by 5′-deoxyadenosylcobalamin in the d-lysine 5,6-aminomutase. Because of the analogies and differences between the these enzymes, we wish to study the d-lysine 5,6-aminomutase mechanism in more detail. As a first step, we have cloned the DNA encoding the enzyme subunits into Escherichia coli, sequenced this DNA, and heterologously expressed active, soluble aminomutase. We here report the cloning, sequence information, and initial characterization of recombinant d-lysine 5,6-aminomutase. Tryptone and yeast extract were from Difco forClostridium medium or Acumedia (Baltimore) and Marcor Development Corp. (Hackensack, NJ), respectively, for E. coli medium. Genomic DNA was isolated with a Genomic Tip from Qiagen, and plasmid DNA was prepared for automated fluorescent sequencing with the Qiagen miniprep kit. Competent JM109 cells (>108 colony forming units/μg), plasmid Maxiprep kits, all plasmid vectors, marker ladders, and most restriction enzymes were from Promega Corp. (Madison, WI). HindIII was from Amersham Pharmacia Biotech, SpeI was from Amersham Pharmacia Biotech or New England Biolabs, and NdeI was purchased from Roche Molecular Biochemicals. Cloned Pyrococcus furiosus DNA polymerase was from Stratagene. Competent DH5αMCR and BL21(DE3)E. coli cells were from Life Technologies, Inc. GeneScreen cationic nylon membranes were from NEN Life Science Products. Polyvinyldifluoride membranes were from Bio-Rad. Amido Black stain was from Sigma. All chemicals were of molecular biology grade or higher. C. sticklandiiwas obtained as freeze-dried cells from the American Type Culture Collection (ATCC, Manassas, VA). After reconstitution, cells were grown anaerobically in medium consisting of, per liter: 4.5 g tryptone, 4.5 g of yeast extract, 1.4 g of KH2PO4, 1.0 g of l-lysine·HCl, 25 mg of CoCl2·H2O, 10 mg of CaCl2·2H2O, 0.20 g of MgSO4·7H2O, 0.9 g of KOH, and either 0.5 g of sodium thioglycolate + 5 mm sodium dithionite (cultures < 60 liters) or 20 mg of sodium dithionite (60 liter culture). A trace of methylene blue was included as a redox indicator. Cells were propagated in 100 ml of anaerobic bottle cultures, and a fresh 100-ml culture used to inoculate a 3-liter starter culture. The overnight 3-liter culture was used to inoculate a 60-liter fermentor at the University of Wisconsin, Department of Biochemistry Pilot Plant. The culture was grown to an A 600 of 0.4, harvested through a Sharples continuous-flow centrifuge, and frozen in liquid nitrogen. Protein concentrations were determined by the method of Lowry et al. (10Lowry O.H. Rosebrough N.J. Farr A.L. Randall R.J. J. Biol. Chem. 1951; 193: 265-275Abstract Full Text PDF PubMed Google Scholar) using bovine serum albumin as a standard. Denaturing polyacrylamide gels were stained with Coomassie Blue. Clostridial 5,6-aminomutase was purified by a modification of the method of Baker et al. (5Baker J.J. van der Drift C. Stadtman T.C. Biochemistry. 1973; 12: 1054-1063Crossref PubMed Scopus (42) Google Scholar). All manipulations were under subdued light, and vessels containing protein were covered with aluminum foil or black plastic to protect against AdoCbl photolysis. One hundred grams of frozen C. sticklandii were thawed in 20 mm KPO4, 2 mm Na2EDTA, pH 7.2, to which 100 μl of 0.5 m phenylmethylsulfonyl fluoride/acetone was added. After centrifuging 20 min at 6,000 ×g at 4 °C, the rinsed pellet was suspended in 20 mm KPO4, pH 7.2, 1 mmdithiothreitol, 1 mm MgCl2, then 50 μl of 0.5m phenylmethylsulfonyl fluoride/acetone, 10 mg of lysozyme, and several crystals of deoxyribonuclease I (Sigma) were added. The cells were incubated 30 min at 37 °C, then 1 ml of 0.2 mNa2EDTA was added and the suspension sonicated once for 1 min at high power with a Heat Systems Ultrasonics, Inc. Sonicator. After centrifugation, the supernatant fluid was dialyzed twice for 3–4 h against 1 liter of 20 mm KPO4, pH 7.2, 1 mm dithiothreitol, 2 mm Na2EDTA, hereafter referred to as standard buffer. The dialyzed solution was loaded onto a 5 × 30-cm DE52 column equilibrated in standard buffer and attached to a Amersham Pharmacia Biotech FPLC system. The column was washed with 600 ml of standard buffer, 0.1 mNaCl and 600 ml of standard buffer, 0.2 m NaCl, and the enzyme was eluted with a 1.2-liter linear (0.3–0.5 m NaCl) gradient into 10-ml fractions. Fractions that were visibly pink or orange during brief light exposure were assayed as described (3Morley C.G.D. Stadtman T.C. Biochemistry. 1970; 9: 4890-4900Crossref PubMed Scopus (43) Google Scholar) with minor modifications (see below) and analyzed by SDS-PAGE. Pooled fractions were concentrated to 500 μl by ultrafiltration, loaded onto a Hi-Prep 26/60 Sephacryl S-200 HR column (Amersham Pharmacia Biotech), and eluted with standard buffer, 0.1 m NaCl into 2-ml fractions. Fractions pooled based on activity and gel purity were loaded onto a Mono Q HR 16/10 column (Amersham Pharmacia Biotech) equilibrated in standard buffer. A 0–1 m NaCl gradient was run at 5 ml/min, and 3-ml fractions were collected. Fractions were pooled based on color and SDS-PAGE purity, concentrated to 7 mg/ml, aliquoted into 0.5-ml Eppendorf tubes, frozen in liquid nitrogen, and stored at −80 °C. To determine N-terminal amino acid sequences of the two polypeptide chains constituting d-lysine 5,6-aminomutase, 14 μg of purified enzyme was separated by SDS-PAGE on a 4–20% acrylamide gradient gel with 0.1% sodium thioglycolate in the electrophoresis buffer. The gel and a polyvinylidene difluoride membrane were prepared, and polypeptides transferred to polyvinylidene difluoride by electrotransfer at 150 mA for 2.75 h. The membrane was stained for 30 s with Amido Black, partially destained, the desired bands thereon delineated, and then destained as completely as possible overnight in deionized water. The bands were excised and sent to the Michigan State Macromolecular Structure, Sequencing, and Synthesis Facility for N-terminal sequencing. Because the order of the subunits in the presumed 5,6-aminomutase operon was not known, polymerase chain reactions were tried usingC. sticklandii genomic DNA and degenerate coding and anticoding primers (Cruachem) designed from the N-terminal information, but without success. To acquire internal peptide sequence information, we used a modified method of Judd (12Judd R.C. Methods Enzymol. 1990; 182: 613-626Crossref PubMed Scopus (17) Google Scholar) to cyanogen bromide digest one of the polypeptide chains and separate the fragments by 20% acrylamide SDS-PAGE. Two bands therefrom were chosen for N-terminal analysis. From the internal sequence, a degenerate anticoding PCR primer (31 kDa3/AC) was designed and used with primer 31/C for subsequent PCR reactions (Table I). A clear ∼500-bp PCR product was excised from agarose gel and ligated into vector pGEM-T. The resulting construct, pCC-4 (Table II), was sequenced from the T7 universal primer site with Sequenase (Amersham Pharmacia Biotech). The predicted amino acid sequence next to the primer sequence matched the internal peptide sequence information.Table IPCR primersPrimer nameSequenceNucleotide31/CATGGARAARAARGARTT1915–193131/ACAAYTCYTTYTTYTCCAT1915–193151/CATGGARAGYAARYTIAAYYTIGAYTTYAA319–34751/ACTTRAARTCIARRTTIARYTTRCTYCCAT319–34731kD3/ACTTYTTDATRAARTCYTCRTT2398–2417NdeIPCRGGAGTGATACATATGGAAAG307–326BamHIPCRAGGATCCATTTCAAAAGCAT1331–1350Bam2PCRTTGAAATGGATCCTGAACTAAAAAATG1337–1363Bam3PCRCAAAGATATATCGGATCCAAATTAAAG2685–2711 Open table in a new tab Table IIPlasmid construct and strain descriptionsPlasmid or strainDescriptionpCC-4pCC-4pGEM-T/500-bp PCR product from C. sticklandii genomic DNApCC-9pGEM-3Zf(−)/5-kbp probe-positive HindIII fragment from C. sticklandii genomic DNApCC-20PCR-amplified upstream portion of kamDE between NdeI mutagenic site and pre-existing internal BamHI site, cloned into SmaI site of pGEM-3Zf(−)pCC-24PCR-amplified downstream portion of kamDE between internal BamHI site and extreme 3′ mutagenic BamHI site, cloned into SmaI site of pGEM-3Zf(−)pCC-28pET-9a containing PCR fragment 3 (pCC-20 insert) and PCR fragment 4 (pCC-24 insert) in proper 5′ → 3′ orientation for KamDE expressionCC-9DH5αMCR cells containing pCC-9CC-10Independent isolate of CC-9CC-11Independent isolate of CC-9CC-28pCC-28 in BL21(DE3) cellsCC-29pET-9a in BL21(DE3) cells Open table in a new tab The pCC-4 insert was gel-purified and used as a template for probe synthesis with [32P]dATP (13Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual, 3. 2nd Ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1989Google Scholar).HindIII-digested Clostridial genomic DNA was Southern hybridized against this probe, and an approximate 5-kbp band was detected by exposure to Kodak Bio-Max film. DNA in this size range was excised from another gel and ligated intoHindIII-digested pGEM-3Zf(−). The resulting construct (pCC-9) was transformed into DH5αMCR cells. In situhybridization on GeneScreen membranes with 32P-labeled probe led to three independent isolate strains: CC-9, CC-10, and CC-11. Thermocycle double-stranded DNA sequencing was performed using materials and protocols from the University of Wisconsin Biotechnology Center. Reactions were run through a Amersham Pharmacia Biotech G-50 spin column, dried, and submitted for analysis. Some convenient deletions next to the T7 universal priming site were made by restriction/religation, and coding strand sequence was determined by sequencing the deleted plasmids from the T7 site or primer walking with pCC-9 as template. DNA sequences were verified by sequencing the anticoding strand of pCC-10 and pCC-11. Sequence information was manipulated with the Wisconsin Package, version 7 (14Genetics Computer Group, I.7.0 Ed. Genetics Computer Group, Inc., Madison1991Google Scholar). The coding regions for the subunits ofd-lysine 5,6-aminomutase, defined as the isolated open reading frames beginning with the experimentally determined N-terminal amino acid sequences, were PCR amplified from pCC-9 in two pieces using high fidelity Pfu polymerase. Mutagenic primer NdeIPCR introduced an NdeI site at the first codon of the large subunit and simultaneously changed the native GTG initiation codon to ATG. The reverse PCR primer BamHIPCR was designed around an internalBamHI site in the Clostridial DNA. The resulting amplified DNA was cloned into the SmaI site of pGEM-3Zf(−) to produce pCC-20. The second half of the coding DNA was amplified between coding primer Bam2PCR and mutagenic anticoding primer Bam3PCR that introduced a BamHI site near the 3′ end of the small subunit coding DNA. This PCR product was also cloned into pGEM-3Zf(−) as before to produce pCC-24. The expression construct was made by ligating theNdeI-BamHI fragment of pCC-20 into theNdeI-BamHI site of pET-9a, isolating the resulting construct in DH5αMCR, then cloning the BamHI insert from pCC-24 into the BamHI site of this intermediate construct and checking for correct orientation by restriction digestion. The entire insert was sequenced to check for PCR-induced mutations, then the complete plasmid was transformed into BL21(DE3) to produce strain CC-28. A control strain CC-29 was made by transforming BL21(DE3) with pET-9a. To test induction properties, 2 ml of LB + 30 μg/ml kanamycin sulfate (LBK) was inoculated with either CC-28 or CC-29 and grown 12 h. Each culture was induced with 12 mg of d-lactose and grown for another 2 h. Samples were taken at 0, 30, 60, 90, and 120 min after induction for gel analysis (Fig. 1). Assays of recombinant enzyme samples initially consisted of 100 mm NaEPPS, pH 8.5, 5 mm MgCl2, 40 mm NH4Cl, 50 mmβ-mercaptoethanol, 40 μm PLP, 20 μmadenosylcobalamin, and 20 mmd-lysine·HCl in a final volume of 50 μl. This composition differed from that reported by Morley and Stadtman (3Morley C.G.D. Stadtman T.C. Biochemistry. 1970; 9: 4890-4900Crossref PubMed Scopus (43) Google Scholar) in the replacement of Tris by EPPS buffer to eliminate the potential reactivity of neutral Tris. In addition, [Mg2+] was made equal to [ATP], and 8 mmdithiothreitol was replaced by 50 mm β-mercaptoethanol. As characterization developed, conditions were modified to maintain saturation with essential components while eliminating non-essential ones. The final standard assay composition was 100 mmNH4EPPS, pH 8.5, 5 mm β-ME, 40 μm PLP, and 100 μm AdoCbl. Acid-quenched assays were analyzed by thin-layer chromatography of radiolabeled substrates and products (3Morley C.G.D. Stadtman T.C. Biochemistry. 1970; 9: 4890-4900Crossref PubMed Scopus (43) Google Scholar). A 1-liter culture of CC-28 was produced by gentle shaking (≤ 200 rpm in an unbaffled, 1-liter Erlenmeyer flask) for 16 h at 37 °C. Cells were harvested by centrifugation for 20 min at 5,000 ×g, 4 °C, and lysed by suspension and sonication (4 × 30 s at medium power with a Fisher Model 550 Sonic Dismembrator) in 10 ml of 10 mm Tris-Cl, pH 7.5, 1 mm EDTA, 0.1% Triton X-100, 0.2 mm NaCl, 1 mm PLP. Phenylmethylsulfonyl fluoride was added to 1 mm, then cell debris pelleted at 5,000 ×g. Dilute polyethyleneimine was added to make the protein solution 0.1% polyethyleneimine in 15 ml, and the solution was centrifuged at 11,000 × g to produce a clarified extract. This extract was diluted to 250 ml with buffer and loaded onto a 2.5 × 30-cm DE52 column equilibrated to constant pH in 20 mm triethanolamine-HCl, pH 7.2, 1 mm β-ME, 10 μm PLP. A 500-ml linear 0–0.5 m NaCl gradient was run and 10-ml fractions collected. Fractions with peak activity toward the end of the gradient were collected, concentrated over a YM30 membrane in an Amicon pressure cell, partially degassed, then loaded onto a Pharmacia Hi-Prep S-200 column equilibrated in the starting buffer for the DE52 column. Fractions (5 ml) were collected, and those with peak activity pooled, concentrated, and stored at −80 °C after flash-freezing aliquots in liquid nitrogen. All glassware for metal analysis was acid-washed and rinsed with freshly deionized 18.3 megohm water. Buffers were treated by passage through biotechnology grade Chelex-100 resin (Bio-Rad). To one of two 300-ml batches of triethanolamine-HCl buffer, pH 7.2, 1 mm β-ME were added α,α′-dipyridyl, o-phenanthroline, Na2EDTA, and oxalic acid to 1 mm each. Two 2.2-mg aliquots (150 μl each) of KamDE 2The large (α) and small (β) subunits have been named KamD and KamE, respectively. The aminomutase complex is referred to as KamDE. A GenBank™ search using the full amino acid sequences of KamD or KamE individually failed to yield any matching polypeptides. were dialyzed overnight against each buffer in 0.5-ml dialysis cassettes (Pierce) at 4 °C. Metal content of the enzyme preparations was analyzed by inductively coupled plasma-mass spectrometry at the University of Wisconsin Soil Science Center. The thermal stability of the dialyzed enzyme was examined in the presence or absence of 140 μm AdoCbl and PLP over the temperature range 37–60 °C. A solution of assay buffer, PLP, and AdoCbl was demetallated by extraction with dithizone/CHCl3as described (15Riordan J.F. Vallee B.L. Methods Enzymol. 1988; 158: 3-6Crossref PubMed Scopus (18) Google Scholar). For thermal stability tests, 2.5-μl aliquots of 4 μm enzyme tetramer were either heated for 5 min or after combination with cofactors (0.6 μm tetramer). The heated samples were kept on ice until assays were performed. KamDE (30 μm tetramer) was incubated at 37 °C, pH 8.5, with 50 mm β-ME, 60 μm PLP, and 100 μm AdoCbl, and either water or 20 mmd-lysine·HCl in a 90-μl volume. Timed aliquots were assayed for 2 min with 200 mmd-lysine·HCl. The inactivation and assays were carried out both aerobically and anaerobically to explore the effect of dissolved oxygen on the inactivation. Anaerobicity was established by performing the experiment in an anaerobic chamber and inclusion of ammonium protocatechuate and protocatechuate 3,4-dioxygenase. This solution was preincubated for 1 h before lysine addition. [15N-DMB]AdoCbl was a gift from Prof. János Rétey. Isotopomer (250 μm) was photolyzed anaerobically by incubation overnight under fluorescent lighting to obtain an enzymatically unbound reference sample. To generate enzyme-bound cob(II)alamin, 64 μm KamDE tetramer was incubated with 100 mm NH4EPPS, pH 8.5, 64 μm [15N-DMB]AdoCbl, 100 μmPLP, and 100 mm methylhydrazine anaerobically in the dark for 30 min at 37 °C. Methylhydrazine was empirically found to generate a well resolved cob(II)alamin spectrum anaerobically without an interfering organic radical near g = 2.0. The sample was transferred to an EPR sample tube and rapidly frozen by immersion into liquid isopentane near liquid nitrogen temperature. EPR spectra were acquired on a Varian E-3 X-band spectrometer with samples held at 77 K by immersion in a cold finger Dewar filled with liquid nitrogen. Relevant spectroscopic parameters were ν = 9.09 GHz, power = 12.5 mW, modulation amplitude = 12.5 G, and time constant = 0.3 s. The nucleotide sequence of the C. sticklandii genomic region encoding the subunits ofd-lysine 5,6-aminomutase is shown in Fig. 2. The sequence of the cloned 5-kbpHindIII fragment of the C. sticklandii genome shows the 5,6-aminomutase coding region to be at the end of an operon. The large (α) subunit of the enzyme is encoded upstream of the small (β) subunit, and is preceded by the 3′ end of an unknown open reading frame. Immediately downstream of the small subunits translational termination codon are inverted repeat sequences that we postulate to be the transcriptional terminator for the operon. A potential Shine-Delgarno sequence GAGGAG similar to the E. coliconsensus AGGAGG is present at (−9)-(−14) relative to the first coding nucleotide of the large subunit; the same region upstream of the small subunit reads GGGAGG, and is also likely to be a ribosome-binding site.FIG. 2Nucleotide sequence and translation of theC. sticklandii d-lysine 5,6-aminomutase genes and surrounding region. Underlined regions are, in 5′ → 3′ order: the putative ribosome-binding sites ford-lysine 5,6-aminomutase α- and β-subunits, inverted repeats for transcriptional termination stem-loop structure, and (−35) and (−10) transcriptional elements for different operon downstream of the genes.View Large Image Figure ViewerDownload (PPT)FIG. 2Nucleotide sequence and translation of theC. sticklandii d-lysine 5,6-aminomutase genes and surrounding region. Underlined regions are, in 5′ → 3′ order: the putative ribosome-binding sites ford-lysine 5,6-aminomutase α- and β-subunits, inverted repeats for transcriptional termination stem-loop structure, and (−35) and (−10) transcriptional elements for different operon downstream of the genes.View Large Image Figure ViewerDownload (PPT) The polypeptide molecular masses for the 5,6-aminomutase subunits are 57,261 Da and 29,191 Da, comparable to the M r values of 55,000 and 30,000 estimated by SDS-PAGE. Codon usage differs from that in E. coli, reflecting the AT-rich Clostridial genome; however, expression of the aminomutase polypeptides in E. coli is efficient despite this difference. The N-terminal amino acid sequence of the small subunit isolated from Clostridiumbegins with serine, implying that the initiatorN-formylmethionine residue of this subunit is removed during post-translational processing in C. sticklandii. Both subunits have excess Glu and Asp relative to Arg, Lys, and His, which rationalizes the acidic character of the enzyme as observed on ion exchange columns at pH 7.5. Comparison of the 5,6-aminomutase subunit amino acid sequences to GenBank™ entries reveals a region in KamE with similarity to B12-dependent mutases, includingClostridial glutamate mutase, Archaeoglobusmethylmalonyl-CoA mutase, and 2-methyleneglutarate mutase fromClostridium barkerii (Fig. 3). Notably, the conserved histidine of these "base-off" mutases that replaces the 5,6-dimethylbenzimidazole bottom axial ligand of adenosylcobalamin upon binding is present in the conserved GXDXHXXG motif. This putative adenosylcobalamin binding sequence correlates with previous observations that the small subunit of the aminomutase binds corrinoids (6Baker J.J. Stadtman T.C. Dolphin D. B12. 2. John Wiley & Sons, Inc., New York1984: 203-231Google Scholar). The large subunit possesses a distinctive group of three cysteine residues in the motif Cys-X 2-Cys-X 3-Cys (C234α, C237α, C241α). This sequence may be compared with the Cys-X 2-Cys-X 4-Cys found ligating a [4Fe-4S] cluster in the corrinoid/iron-sulfur protein (16Lu W.-P. Schiau I. Cunningham J.R. Ragsdale S.W. J. Biol. Chem. 1993; 268: 5605-5614Abstract Full Text PDF PubMed Google Scholar) and the structural zinc binding motif Cys-X 2-Cys-X 2-Cys in alcohol dehydrogenase (17Vallee B.L. Falchuk K.H. Physiol. Rev. 1993; 73: 79-118Crossref PubMed Google Scholar). However, the spacing between the second and third cysteines in KamD differs from both. Unlike l-lysine 2,3-aminomutase, KamDE does not contain iron or inorganic sulfide, nor does it display spectral characteristics of iron-sulfur centers. The significance of these grouped thiols, if any, is unknown. As seen in Fig. 1, CC-28 continuously expresses the 5,6-aminomutase polypeptides from the pET vector construct during vigorous aerobic growth at 37 °C. These peptides were found entirely as inclusion bodies. Under less vigorous agitation, however, a significant portion of these peptides are present as soluble, active enzyme in a cell-free extract (Fig. 4). A liter of LBK culture usually provides ≥50 mg of purified, soluble apo-B12 KamDE with a specific activity greater than that seen with enzyme purified from Clostridium (Table III).Table IIID-Lysine 5,6-aminomutase purification tableFractionVolumeProteinActivitySpecific activityPurificationYieldmlmgIUIU/mg-fold%Crude13.54589402.051100PEI SN14.33964801.211.16DE52903778942.371.1695S-200 + amicon1.4221978.954.3721 Open table in a new tab Allosteric regulation by ATP was seen in early studies of the 5,6-aminomutase (3Morley C.G.D. Stadtman T.C. Biochemistry. 1970; 9: 4890-4900Crossref PubMed Scopus (43) Google Scholar). However, as shown Fig. 5, 5 mm ATP does not significantly affect lysine saturation properties of the recombinant KamDE. Any effects seen at higher concentrations would have questionable mechanistic significance in vivo. Unlike what was seen for the combined E1-E2 system, no change from sigmoidal to hyperbolic kinetics occurs for the purified recombinant enzyme. The minor enhancement ofV max with added MgATP in Fig. 5 was not observed consistently in subsequent studies, and is not considered significant. The purified enzyme contains a substantial quantity of metal ions, primarily Zn2+. When dialyzed against a mixture of chelators, however, the metal content was lowered to ≤4% of tetramer present. The activities and thermal stabilities of untreated and metal-depleted KamDE were compared using metal-free reagents (excepting the Co2+ in AdoCbl). As shown in Fig. 6, almost no activity difference was seen between the two preparations with 37 °C preincubation, minimizing the possibility of an essential catalytic metal. Chelation also does not result in greater temperature sensitivity, as might be observed upon removal of a structural metal; in fact, the chelated enzyme shows slightly greater temperature stability. While not definitive evidence against the presence of a structural metal in KamDE in vivo, the similarity in thermal lability between as purified and metal-depleted KamDE does not support such presence. The 5,6-aminomutase purified from C. sticklandii was found to undergo rapid inactivation in the absence of the activating enzyme E 2 + ATP (6Baker J.J. Stadtman T.C. Dolphin D. B12. 2. John Wiley & Sons, Inc., New York1984: 203-231Google Scholar). To examine inactivation of the recombinant enzyme,
Referência(s)