The glyoxylate bypass of Ralstonia eutropha

FEMS Microbiology Letters 228 (2003) 63^71

www.fems-microbiology.org

The glyoxylate bypass of Ralstonia eutropha Zheng-Xiang Wang a

a;b

, Christian O. Bra«mer a , Alexander Steinbu«chel

a;

Institut fu«r Molekulare Mikrobiologie und Biotechnologie, Westfa«lische Wilhelms-Universita«t Mu«nster, CorrensstraMe 3, 48149 Mu«nster, Germany b Research Centre of Industrial Microbiology, School of Biotechnology, Southern Yangtze University, 214036 Wuxi, PR China Received 20 July 2003; received in revised form 16 September 2003; accepted 17 September 2003 First published online 3 October 2003

Abstract The glyoxylate bypass genes aceA1 (isocitrate lyase 1, ICL1), aceA2 (isocitrate lyase 2, ICL2) and aceB1 (malate synthase, MS1) of Ralstonia eutropha HF39 were cloned, sequenced and functionally expressed in Escherichia coli. Interposon-mutants of all three genes (vaceA1, vaceA2 and vaceB1) were constructed, and the phenotypes of the respective mutants were investigated. Whereas R. eutropha HF39vaceA1 retained only 19% of ICL activity and failed to grow on acetate, R. eutropha HF39vaceA2 retained 84% of acetate-inducible ICL activity, and growth on acetate was not retarded. These data suggested that ICL1 is in contrast to ICL2 induced by acetate and specific for the glyoxylate cycle. R. eutropha HF39vaceB1 retained on acetate as well as on gluconate about 41^42% of MS activity and exhibited retarded growth on acetate, indicating the presence of a second hitherto not identified MS in R. eutropha HF39. Whereas in R. eutropha HF39vaceA1 and R. eutropha HF39vaceA2 the yields of poly(3-hydroxybutyric acid), using gluconate as carbon source, were significantly reduced, R. eutropha HF39vaceB1 accumulated the same amount of this polyester from gluconate as well as from acetate as R. eutropha HF39. : 2003 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. Keywords : Glyoxylate bypass; Isocitrate lyase ; Malate synthase; Polyhydroxyalkanoate; Ralstonia eutropha

1. Introduction Most microorganisms utilizing acetate or other fatty acids as sole carbon source employ the glyoxylate cycle for biosynthesis of precursors of cell compounds. The glyoxylate cycle bypasses the decarboxylation steps of the tricarboxylic acid cycle (TCA), thereby preventing complete oxidation of acetyl^CoA to carbon dioxide. By generating oxaloacetate as acceptor for acetyl^CoA, it permits net incorporation of carbon into cellular metabolic £uxes [1]. Five enzymes are involved in this cycle, three of them are the TCA cycle enzymes citrate synthase (4.1.3.7), aconitase (4.2.1.3) and malate dehydrogenase (1.1.1.37); the other two enzymes, isocitrate lyase (ICL, EC 4.1.3.1) and malate synthase (MS, EC 4.1.3.2) are unique to the glyoxylate cycle. ICL cleaves isocitrate to succinate and glyoxylate, and MS condenses glyoxylate with acetyl^CoA to yield malate. With each turn of the cycle 2 mol of acetyl^ CoA is introduced, resulting in the synthesis of 1 mol of

the four-carbon compound oxaloacetate [1]. The availibility of acetyl^CoA and NADPH is in Ralstonia eutropha most important for biosynthesis of poly(3-hydroxybutyric acid) (poly(3HB)) and other polyhydroxyalkanoates (PHAs), and their intracellular concentrations signi¢cantly a¡ect the amount and also the composition of PHAs [2]. Acetyl^CoA and NADPH are mainly generated by the Entner^Doudoro¡ pathway, the pentose phosphate cycle, the pyruvate dehydrogenase complex and the TCA cycle. On carbon sources metabolized primarily to acetyl^CoA, the glyoxylate bypass plays an important role for supply of substrates for PHA synthesis. In this study, we describe the cloning, disruption and functional analysis of genes aceA1, aceA2 and aceB1 encoding ICL isoenzyme 1 (ICL1), ICL isoenzyme 2 (ICL2) and malate synthase 1 (MS1) of R. eutropha HF39.

2. Materials and methods 2.1. Bacteria, plasmids and cultivation conditions

* Corresponding author. Tel. : +49 (251) 8339821; Fax : +49 (251) 8338388. E-mail address : [email protected] (A. Steinbu«chel).

All strains and plasmids investigated are listed in Table 1. R. eutropha HF39 and its derivates were grown at

0378-1097 / 03 / $22.00 : 2003 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/S0378-1097(03)00722-5

FEMSLE 11240 30-10-03

64

Z.-X. Wang et al. / FEMS Microbiology Letters 228 (2003) 63^71

30‡C in nutrient broth (NB; 0.8%, w/v) or in mineral salts medium (MM) [3] supplemented with ¢lter-sterilized carbon sources plus 500 Wg ml31 streptomycin. Escherichia coli was cultivated at 37‡C in Luria^Bertani (LB) medium [4]. When necessary, ampicillin (75 Wg ml31 ), kanamycin (50 Wg ml31 ), chloramphenicol (34 Wg ml31 ) or tetracycline (25 Wg ml31 ) was added to the medium.

2.4. Polymerase chain reaction (PCR) ampli¢cations PCR ampli¢cations of DNA were carried out as described in a manual [4] using PLATINUM0 Pfx-DNA polymerase (Gibco BRL Life Technologies, Karlsruhe, Germany) and the primers shown in Table 2. 2.5. In situ DNA hybridization

2.2. Isolation and manipulation of nucleic acids DNA manipulations and isolation of plasmid DNA were done according to [4] and enzymes were used according to the instructions of the manufacturers. For preparation of cosmid DNA a NucleoBond BAC Maxi kit (Clontech Laboratories, USA) was used. R. eutropha HF39 genomic DNA was isolated according to [5] with the following modi¢cations: cells from a 100-ml culture were harvested by centrifugation and incubated at 42‡C in the presence of sodium dodecyl sulfate (1%, w/v), RNase A (25 Wg ml31 ) and proteinase K (0.5 Wg ml31 ) until the solution became opaque. The supernatant was collected and dialyzed against distilled water. DNA fragments were isolated from agarose gels using a Nucleotrap kit (Macherey-Nagel, Du«ren, Germany). 2.3. Transfer of DNA Competent cells of E. coli were prepared and transformed by the CaCl2 procedure [4]. For transduction of R. eutropha HF39 genomic DNA to E. coli S17-1, DNA was ligated into cosmid pHC79 and packaged by an in vitro packaging kit (Stratagene, USA). DNA transfer into R. eutropha HF39 was achieved by conjugation [6].

DNA probes were obtained from genomic DNA of R. eutropha HF39 by PCR based on highly conserved regions from related nucleotide sequences of R. metallidurans CH34 after alignment of the sequences of aceA1, aceA2 or aceB1 from Ralstonia metallidurans CH34, Ralstonia solanacearum, Mycobacterium tuberculosis and Pseudomonas aeruginosa. Oligonucleotides aceAP1-1 and aceAP1-2, aceAP2-1 and aceAP2-2 or aceBP1-1 and aceBP1-2 were used to amplify aceA1, aceA2 or aceB1, respectively. The 740-bp (aceA1740 ), the 1280-bp (aceA21280 ) and the 640-bp (aceB1640 ) PCR products were puri¢ed and inserted into pGEM-T-Easy (Promega, USA) to yield pGEMaceA1P and pGEMaceB1P, or into pBluescript SK3 to yield pSKaceA2P. Colony hybridization was done at 68‡C using digoxigenin (DIG)-labeled aceA1740 , aceA21280 or aceB1640 as probes [7] and alkaline phosphatase anti-DIG antibodies plus NBT/BCIP (Roche Diagnosis, Germany). 2.6. DNA sequence analysis The primer-hopping strategy [8] was applied to determine nucleotide sequences using the Sequi Therm EXCEL TM II long-read cycle sequencing kit (Epicentre Tech-

Table 1 Bacterial strains and plasmids Strains or plasmids Strains R. eutropha HF39 VG12 HF39vaceA1 HF39vaceA2 HF39vaceB1 E. coli XL1-Blue S17-1 CGSC 5233 GSC 5234 Plasmids pHC79 pBluescript SK3 pSKsym6Km pSUP202 pBBR1MCS

Relevant characteristics

Reference and source

Sm-resistant derivative of R. eutropha H16 (wild-type) Tn5-induced mdh3 mutant of R. eutropha HF39 HF39vaceA1: :6Km HF39vaceA2: :6Km HF39vaceB1: :6Km

[22] [11] This study This study This study

recA1, endA1, gyrA96, thi1, hsdR17, supE44, relA1, V3 , lac [FP, proAB, lacIq , lac ZvM15, Tn10(Tcr )] thi1, proA, hsdR17, recA1, tra(RP4) thr-1, leuB6, fhuA2, lacY1, glnV44(AS), gal-6, V3 , hisG1(Fs), malT1(VR ), xylA7, mtlA2, ppc-2, thi-1, aceA7 E. coli thr-1, leuB6, fhuA2, lacY1, glnV44(AS), gal-6, V3 , hisG1(Fs), malT1(VR ), xylA7, mtlA2, ppc-2, aceB6 E. coli

[23]

Tcr , Apr , cosmid Apr , lacPOZP 6Km in pSKSym Tcr , Apr , Cmr , mob Cmr , mob

[25] Stratagene (CA, USA) [26] [24] [27]

FEMSLE 11240 30-10-03

[24] Genetic Stock Center (New Haven, CT, USA) Genetic Stock Center (New Haven, CT, USA)

Z.-X. Wang et al. / FEMS Microbiology Letters 228 (2003) 63^71

65

Table 2 Primers used for PCR Primer

Nucleotide sequence

Usage

aceAP1-1 aceAP1-2 aceAP2-1 aceAP2-2 aceBP1-1 aceBP1-2 aceA1-5up 3PaceA-down aceA1(EcoRI) aceA2(EcoRI) A2up A2dd aceB-up aceB-3inv Km-1 Km-2 aceA1 (F) aceA1 (R) aceA2 (F) aceA2 (R) aceB1 (F) aceB1 (R)

5P-GCCAATGGTCGTGAAGCGCATCAACAACAC-3P 5P-CCGTGGGCCAGGTTGAACATCGTGTAGTTC-3P 5P-GCGGAATTCGACATCAAGGCAGTTGCTGGTTTGAA-3P 5P-GCGGAATTCGGCCTGGAACGTGCGGATCTTCTCGT-3P 5P-GTGCTGATCGAGACGATCCTGGCCGCGTTC-3P 5P-ATCCACTGCCACACCTGGGAGCGCGAGATC-3P 5P-GTGTTGTTGATGGGCTTGACGACTTGC-3P 5P-TCTTCGAAGACAAGAAAGTGGCCTGA-3P 5P-GCGGAATTCCAGAACAGATCCAGGCCCTGC-3 5P-GCGGAATTCTGAAACACCCGTTTGAACAGC-3P 5P-GTGTACTTCGACGAGTCGGCATC-3P 5P-GATCAGACCGAACTGGCGAAGCTCG-3P 5P-CGAATCCTCGAAGTCGGTCA-3P 5P-CAACGTCGGCATCCACTACCTG-3P 5P-TGAACAAGATGGATTGCACGCAGG-3P 5P-TCAAGAAGGCGATAGAAGGCGATGC-3P 5P-GTGGATCCGACGGTTGCCATGGTGTGCG-3P 5P-GTGGGATCCAATCAGACGATTCCTAATTACGC-3P 5P-GTGAATTCTTATTGACCGACTCTGGCCG-3P 5P-GTGAATTCATGGAAGCGCACAAGAATATGTT-3P 5P-GCCCTCGAGATGCTCTTGTCTCGTCCAACG-3P 5P-CTGCTCGAGCCC CGTATTGCGAAGATTGGATT-3P

aceA1 aceA1 aceA2 aceA2 aceB1 aceB1 aceA1 aceA1 aceA1 aceA1 aceA2 aceA2 aceB1 aceB1 6Km 6Km aceA1 aceA1 aceA2 aceA2 aceB1 aceB1

nologies, WI, USA), IRD800-labeled oligonucleotides (MWG-Biotech, Ebersberg, Gemany) and a LI-COR 4000L automatic sequencing apparatus (MWG-Biotech, Ebersberg, Germany). Sequence data analysis and homology searches were performed using the FASTA program and the BLAST network service (National Centre for Biotechnology Information, France). Nucleotide sequences of aceA1, aceA2 and aceB1 were deposited in the GenBank database under accession nos. AJ421510, AF499030 and AJ421511. 2.7. Deletion mutants of aceA1, aceA2 or aceB1 of R. eutropha HF39 Deletion mutants were screened based on replacement of the central regions of the structural genes by a kanamycin-resistance gene (6Km) [6]. For deletion of aceA1, oligonucleotides aceA1-5up and 3PaceA-down were used to remove part of aceA1 by inverse PCR employing pSKaceA1 as template. The PCR product was ligated with 6Km, resulting in pSKaceA1P: :6Km. With aceA1(EcoRI) and aceA2(EcoRI) aceA1P: :6Km was ampli¢ed by PCR using pSKaceA1P: :6Km as template. The PCR product was ligated into EcoRI-restricted pSUP202 to retrieve pSUPCm: :aceA1P6Km. For deletion of aceA2, A2up and A2dd were used to remove part of aceA2 by inverse PCR using pSKaceA2 as template. The PCR product was ligated with 6Km yielding pSKaceA2P: :6Km. The aceA2P: :6Km fragment was recovered by digestion of pSKaceA2P: :6Km with EcoRI and ligated into the EcoRI site of pSUP202 to yield pSUPCm : :aceA2P6Km. For deletion of aceB1, aceB-up and aceB-3inv were used to re-

move part of aceB1 by inverse PCR using pSKaceB1 as template. The PCR product was puri¢ed and ligated with 6Km to yield pSKaceB1P: :6Km. The XhoI-digested aceB1P: :6Km cassette was ligated into the SalI site of pSUP202 resulting in pSUPTc: :aceB1P6Km. Subsequently, pSUPCm: :aceA1P6Km, pSUPCm: :aceA2P6Km or pSUPTc: :aceB1P6Km were transferred from E. coli S17-1 to R. eutropha HF39. Transconjugants were selected on NB plates containing 500 Wg streptomycin and 160 Wg kanamycin per ml. Exchanges of functional aceA1, aceA2 or aceB1 with deleted genes by double crossover were selected and con¢rmed by DNA sequencing. 2.8. Expression of the R. eutropha HF39 aceA1, aceA2 and aceB1 in E. coli The complete aceA1, aceA2 or aceB1 genes and adjacent regions were ampli¢ed by PCR. Primers aceA1 (F) and aceA1 (R) containing BamHI sites were used to amplify aceA1 with a 640-bp upstream and a 480-bp downstream sequence. Primers aceA2 (F) and aceA2 (R) containing EcoRI recognition sites were used to amplify aceA2 with a 320-bp upstream and a 100-bp downstream sequence. Primers aceB1 (F) and aceB1 (R) containing XhoI restriction sites were used to amplify aceB1 with a 240-bp upstream and a 300-bp downstream sequence. PCR products were digested with BamHI, EcoRI or XhoI and inserted into correspondingly linearized pBluescript SK3 to yield pSKaceA1, pSKaceA2 or pSKaceB1. pSKaceA1 and pSKaceA2 were transformed into the ICL mutant CGSC 5233 of E. coli (aceA7v), and pSKaceB1 into the MS mutant CGSC 5234 of E. coli (aceB6v).

FEMSLE 11240 30-10-03

66

Z.-X. Wang et al. / FEMS Microbiology Letters 228 (2003) 63^71

2.9. Preparation of cell extracts and assay of enzymes About 0.2 g of wet cells were washed with ice-cold sodium chloride solution (0.85%, w/v) and resuspended in 2 ml of 50 mM morpholinepropane-sulfonic acid (MOPS; pH 7.5) bu¡er containing 10% (v/v) glycerol, 1 mM EDTA, 1 mM benzamidine and 2 mM dithiothreitol. After passage through a French Press Cell (Aminco, USA) and incubation with 50 Wl RNase (10 mg ml31 ) and 50 Wl DNase (10 mg ml31 ) at room temperature, the solution was centrifuged at 27 000Ug and 4‡C for 30 min, and the supernatant was desalted by passing through a PD-10 column (Amersham Pharmacia Biotech AB, USA). ICL activity was measured in 50 mM MOPS bu¡er (pH 7.5) containing 5 mM MgCl2 and 4 mM phenylhydrazine hydrochloride at 25‡C. The reaction was started by addition of 2 mM trisodium D,L-isocitrate, and the formation of the phenylhydrazone derivative of glyoxylate was monitored at 324 nm (O = 14.63 cm2 Wmol31 ). MS activity was determined by monitoring the release of CoA during the enzymatic reaction at 25‡C using 5,5P-dithiobis(2-nitrobenzoic acid) at 412 nm (O = 13.6 cm2 Wmol31 ) in 50 mM HEPES bu¡er (pH 8.0) containing 2 mM EDTA, 100 mM KCl, 0.1 mM 5,5P-dithiobis(2-nitrobenzoic acid), 0.2 mM glyoxylic acid and 0.2 mM acetyl^CoA. One unit of enzyme activity was de¢ned as the conversion of 1 Wmol of substrate per minute Protein concentrations were determined by the method of Bradford [9] with crystalline bovine serum albumin fraction V (Serva Feinbiochemica, Germany) as reference. 2.10. Isolation and analysis of PHA For quantitative determination of PHA, 8^10 mg lyophilized cells were subjected to methanolysis in the presence of 15% (v/v) sulfuric acid at 100‡C for 5 h, and the resulting hydroxyacyl methylesters were analyzed by gas chromatography as described previously [10]. 2.11. Overexpression of aceA1 in R. eutropha VG12 The aceA1 gene with a 640-bp upstream and a 480-bp downstream sequence was isolated from pSKaceA1 and ligated into pBBR1MCS, resulting in pBBRaceA1, which was then transferred into the Tn5-induced mdh mutant VG 12 of R. eutropha HF39 by conjugation [11].

3. Results 3.1. Identi¢cation and characterization of the R. eutropha HF39 ICL genes aceA1 and aceA2 and the MS gene aceB1 To clone the R. eutropha HF39 glyoxylate cycle genes, the PCR products aceA1740 , aceA21280 and aceB1640 were

obtained from R. eutropha HF39 and used as probes to identify the complete genes in cosmid-mediated genomic libraries of R. eutropha HF39 by colony hybridization. Clones harboring a 32-kb (pHCaceA1 ), a 35-kb (pHCaceA2 ) or a 30-kb DNA insert (pHCaceB1 ), which hybridized with aceA1740 , aceA21280 or aceB1640 , respectively, were identi¢ed. The complete sequences of aceA1, aceA2 and aceB1 were revealed using hybrid cosmids pHCaceA1, pHCaceA2 or pHCaceB1 as templates. 3.2. Characterization of the R. eutropha HF39 aceA1 gene A 3545-bp nucleotide sequence of pHCaceA1 was analyzed, and the putative aceA1 gene encoding ICL1 of R. eutropha HF39, which is preceded by a tentative ribosome binding site, was identi¢ed. AceA1 encodes a protein consisting of 431 amino acid residues with a calculated molecular mass of 47 006 Da. The deduced amino acid sequence showed 89.3, 84.6 or 22.8 mol% identity to ICL1 (Icl1p) of R. metallidurans CH34, Icl1p of R. solanacearum and ICL2 (Icl2p) of R. metallidurans CH34, respectively. Analysis of the primary amino acid sequence of ICL1 revealed an ICL signature pattern composed of the amino acid residues K(185), K(186), C(187), G(188), H(189) and M(190) (Fig. 1A), with cysteine187 as putative active site residue [12]. This highly conserved signature pattern is also found in 2-methylisocitric acid lyases (PrpB) which cleaves 2-methylisocitric acid into succinate and pyruvate as the ¢nal step of the methylcitric acid cycle (Fig. 1A) [6]. A second open reading frame (ORF1) encoding a protein composed of 292 amino acids with a calculated molecular mass of 32 580 Da and exhibiting 49 mol% identity to a probable short-chain dehydrogenase PA3324 of P. aeruginosa (http://www.ncbi.nlm.nih.gov/) was identi¢ed 301 bp upstream of aceA1. No further ORF could be identi¢ed in the 849-bp downstream region of aceA1. The arrangement of aceA1 and ORF1 in the genome was identical to that of R. metallidurans CH34 aceA1; however, di¡erent genes were located in the aceA1 downstream region. 3.3. Characterization of the R. eutropha HF39 aceA2 gene A 2521-bp nucleotide sequence of pHCaceA2 revealed the putative aceA2 gene encoding ICL2 of R. eutropha HF39, which is preceded by a tentative ribosome binding site. The translational product of aceA2 comprised 526 amino acid residues with a calculated molecular mass of 57 890 Da. The deduced amino acid sequence was 94.3, 80.0, 22.8 or 23.3 mol% identical to those of R. metallidurans CH34 Icl2p, P. aeruginosa Icl1p, R. solanacearum Icl1p and R. metallidurans CH34 Icl1p, respectively. Analysis of the primary amino acid sequence of R. eutropha HF39 ICL2 exhibited similarities to a putative signature pattern for ICLs composed of the amino acid residues

FEMSLE 11240 30-10-03

Z.-X. Wang et al. / FEMS Microbiology Letters 228 (2003) 63^71

67

Fig. 1. Signature patterns of ICL or methylisocitrate lyases (A) and MS (B). The conserved hexapeptide, K[KR]CGH[LMQR], which can be used as a signature pattern for ICL in A, and the conserved pattern [KR][DENQ]HxxGLNxGxWDY[LIVM]F, which can be used as a signature for MS in B, are shown in bold font. A: (1) R. eutropha HF39 ICL1; (2) R. eutropha HF39 ICL2; (3) R. metallidurans CH34 Icl1p (http://genome.ornl.gov/microbial/reut/); (4) R. metallidurans CH34 Icl2p (http://genome.ornl.gov/microbial/reut/); (5) R. solanacearum Icl1p (NC_003295); (6) D. radiodurans Icl1p (NC_001263); (7) P. aeruginosa Icl1p (NC_002516); (8) E. coli Iclp (X12431) ; (9) M. tuberculosis Icl1p (NC_002755); (10) Corynebacterium glutamicum Icl1p (X75504) ; (11) R. eutropha HF39 PrpB (AAL03988). B: (1) R. eutropha HF39; (2) R. solanacearum (NC_003295); (3) D. radiodurans (NC_001263); (4) E. coli MSA (M19038); (5) Streptomyces coelicolor (AF206498); (6) Vibrio cholerae (NC_002505); (7) S. cerevisiae Mcl1p (NC_001146); (8) Salmonella typhimurium (NC_003197); (9) Aspergillus nidulans (X56671) ; (10) Pichia angusta (P21360); (11) Candida albicans (AF222907); (12) Candida ropicalis (D13415).

K(216), Q(217), C(218), G(219), H(220) and Q(221), whereas the amino acid residues Q(217) and Q(221) are di¡erent from ICL1 and PrpB of R. eutropha HF39 (Fig. 1A). No further ORF could be identi¢ed in the regions upstream (540 bp) or downstream (416 bp) of aceA2.

nacearum was identi¢ed 68 bp upstream of aceB1. No further ORF was found in a region 509 bp downstream of aceB1. The vicinity of aceB1 in the genome of R. eutropha HF39 was similar to that of R. metallidurans CH34.

3.4. Characterization of the R. eutropha HF39 aceB1 gene

3.5. Heterologous expression of R. eutropha HF39 aceA1, aceA2 or aceB1 in E. coli

A 2636-bp nucleotide sequence of pHCaceB1 was analyzed, and the putative aceB1 gene encoding the MS of R. eutropha HF39, which is preceded by a ribosome binding site, was identi¢ed. The translational product consisted of 528 amino acid residues with a calculated molecular mass of 59 085 Da. The deduced amino acid sequence showed 84.6, 70.1, 56.9 or 17.3 mol% identity to MS1 (Mls1p) of R. solanacearum, R. metallidurans CH34 and Deinococcus radiodurans and to MS2 (Mls2p) of D. radiodurans, respectively. The consensus pattern of MS was found in MS1 of R. eutropha HF39 (Fig. 1B). A second incomplete ORF, ORF4P, encoding a putative haloacid dehalogenase-like hydrolase sharing 70.2 mol% identity to a putative 2-haloalkanoic acid dehalogenase of R. sola-

The aceA1 gene including 640 bp of the upstream and 480 bp of the downstream regions as well as the aceA2 gene including 320 bp of the upstream and 100 bp of the downstream regions were inserted into plasmid pBluescript SK3 and subsequently transformed into the E. coli ICL-negative mutant CGSC 5233 (aceA7v). In addition, aceB1 including 240 bp of the upstream and 300 bp of the downstream regions was inserted into plasmid pBluescript SK3 and subsequently transformed into the E. coli MSnegative mutant CGSC 5234 (aceB6v). The measured speci¢c activities of ICL (70 and 19 U g31 , respectively) and of MS (52 U g31 ) after 18 h growth of the transformants in LB medium at 37‡C clearly showed that aceA1, aceA2 and aceB1 were functionally expressed in E. coli.

FEMSLE 11240 30-10-03

68

Z.-X. Wang et al. / FEMS Microbiology Letters 228 (2003) 63^71

3.6. Generation and characterization of R. eutropha HF39 vaceA1, vaceA2 and vaceB1 interposon-mutants Interposon-mutants of the genes aceA1, aceA2 and aceB1 of R. eutropha HF39 were obtained as described, and the mutants R. eutropha HF39vaceA1, R. eutropha HF39vaceA2 and R. eutropha HF39vaceB1 were used for further analysis. For control, genomic DNA of all mutants was digested with HindIII, ligated into HindIIIlinearized pBluescript SK3 vector, and kanamycin-resistant clones were selected. The sequences of both kanamycin cassette £anking regions of R. eutropha HF39vaceA1, R. eutropha HF39vaceA2 and R. eutropha HF39vaceB1 were completely identical to those of aceA1, aceA2 and aceB1, respectively. This indicated that all target genes were correctly deleted in R. eutropha HF39. Activities of ICL and MS in crude extracts of the aceA1, aceA2 or aceB1 interposon-mutants were measured and compared with those of the parent strain. Since growth of the mutants was much di¡erent from that of the parent strain, two-stage cultivation experiments were done. Cells were ¢rst grown in NB medium, harvested and washed and then transferred to MM containing acetate or gluconate. After 4 h cultivation, the cells were harvested and analyzed. In contrast to MS, expression of ICL was signi¢cantly induced by acetate in R. eutropha HF39. R. eutropha HF39vaceA1 grown on acetate revealed only 19% of ICL activity, while R. eutropha HF39vaceA2 retained 84% of the activity of the parent strain (Table 3). In addition, ICL was strongly induced by acetate in R. eutropha HF39vaceA2 suggesting that aceA1 is acetate inducible. Furthermore, the activities of MS in both R. eutropha HF39vaceA1 and R. eutropha HF39vaceA2 were about 20^30% lower (Table 3). However, in R. eutropha HF39vaceB1 grown on either carbon source, MS activity was about 60% lower in comparison to the parent strain R. eutropha HF39 (Table 3). 3.7. Disruption of glyoxylate bypass signi¢cantly a¡ects growth on acetate as sole carbon source To investigate whether mutations of aceA1, aceA2 or

aceB1 a¡ect the catabolism of acetate or gluconate, growth of the interposon-mutants was analyzed (Fig. 2). Cells were cultivated in NB medium, harvested, washed and afterwards resuspended in 50 ml MM supplemented with sodium acetate (0.5%, w/v) or sodium gluconate (0.5%, w/v). Cultivation was carried out at 120 rpm and 28‡C for up to 96 h, and growth was recorded by measuring the increase of turbidity at 600 nm. R. eutropha HF39vaceA1 failed to grow on acetate. In contrast, R. eutropha HF39vaceA2 grew on acetate and gluconate similarly to the parent strain. R. eutropha HF39vaceB1 was able to grow on acetate and gluconate; however, its growth was signi¢cantly reduced on acetate. To determine the possible e¡ect of these mutations on growth on other carbon sources, the growth pattern was also examined using 0.2% (w/v) of the sodium salts of malate, lactate, fumarate, pyruvate, propionate, isocitrate or glyoxylate as sole carbon source. Reduced growth was only found for R. eutropha HF39vaceA1 and R. eutropha HF39vaceB1 when these mutants were cultivated on propionate, indicating that the glyoxylate bypass is in R. eutropha HF39 also required as anaplerotic reaction during the utilization of substrates which are catabolized via the 2-methylcitric acid cycle [11]. 3.8. E¡ect of aceA1, aceA2 or aceB1 disruption on PHA accumulation To examine the e¡ect of mutations of aceA1, aceA2 or aceB1 on accumulation of PHA, R. eutropha HF39 and the respective mutants were grown in MM containing 0.5% (w/v) sodium gluconate or 0.5% (w/v) sodium acetate. The cells in the stationary phase were collected, lyophilized and the PHA content was analyzed by gas chromatography. Both R. eutropha HF39vaceA1 and HF39vaceA2 revealed a signi¢cantly lower poly(3HB) content while it was identical to that of the parent strain R. eutropha HF39 in R. eutropha HF39vaceB1 if gluconate was used as carbon source. During growth on acetate, the poly(3HB) content was not a¡ected in R. eutropha HF39vaceA2 or R. eutropha HF39vaceB1 (Table 4). Surprisingly, overexpression of aceA1 in R. eutropha

Table 3 Speci¢c activities of ICL and MS in crude extracts of R. eutropha HF39 and the mutants R. eutropha HF39vaceA1, HF39vaceA2 and HF39vaceB1 Substrate

ICL Gluconate Acetate MS Gluconate Acetate

Speci¢c activities (U g31 protein) in R. eutropha HF39 and the mutantsa HF39

HF39vaceA1

HF39vaceA2

HF39vaceB1

20.5 (3.5) 73.2 (6.8)

6.6 (2.6) 13.8 (2.1)

4.4 (0.7) 61.3 (10.5)

28.3 (4.5) 73.2 (8.3)

151.5 (21.6) 158.1 (15.0)

120.4 (19.4) 133.1 (25.2)

101.2 (23.1) 110.7 (18.4)

62.5 (12.0) 66.8 (5.8)

a

Cells were cultivated overnight in NB medium at 30‡C and 120 rpm to the early stationary growth phase. The media contained 500 Wg ml31 streptomycin for HF39 or 500 Wg ml31 streptomycin and 160 Wg ml31 kanamycin for mutants. Cells were then collected and washed, and the cell pellets were then suspended in MM supplemented with 0.5% (w/v) sodium gluconate or 0.5% (w/v) sodium acetate and incubated at 30‡C for 4 h with gentle agitation. Data are means of triplicate determinations ( W S.D. in parentheses).

FEMSLE 11240 30-10-03

Z.-X. Wang et al. / FEMS Microbiology Letters 228 (2003) 63^71

69

Fig. 2. Growth behavior of aceA1, aceA2 and aceB1 deletion mutants of R. eutropha HF39. Symbols : vaceA1 mutant on gluconate (a) and on acetate (b); vaceB1 mutant on gluconate (O) and on acetate (R); vaceA2 mutant on gluconate (+) and on acetate (x); R. eutropha HF39 on gluconate (E) and on acetate (F).

VG12, a Tn5 mutant of R. eutropha HF39 defective in the mdh gene encoding a NADH-dependent malate dehydrogenase, did not only improve its growth on gluconate (data not shown) but also restored its poly(3HB) yield from gluconate as substrate (Table 4). Interestingly, R. eutropha VG12 exhibited only about one fourth of the speci¢c activities of ICL, MS and NADPH-dependent isocitrate dehydrogenase. Levels of the wild-type were partly restored in the recombinant mutant harboring pBBRaceA1 (data not shown). An explanation for this cannot be provided.

4. Discussion In microorganisms, genes for ICLs and MSs involved in glyoxylate bypass show a great variability regarding number and organization. Whereas, for example, only one copy of ICL (aceA) and one copy of MS (aceB) exist in the archaeon Haloferax volcanii [13] and also in R. solanacearum [14], two copies for ICL (ICL1 and ICL2) and for MS (MLS1 and MLS2) exist in Saccharomyces cerevisiae [15]. Most microorganisms like D. radiodurans possess only one type of ICL [16], and the occurrence of ICL isoenzymes is rare. If two isoenzymes occur, the function of the second is mostly unknown. Like previously revealed by genome sequencing in M. tuberculosis (http://www. ncbi.nlm.nih.gov/PMGifs/Genomes/eub_g.html), R. metallidurans CH34 (http://genome.ornl.gov/microbial/reut/) and S. cerevisiae (http://genome-www.stanford.edu/Saccharomyces/), two ICL isoenzymes were also found in R. eutropha HF39 in this study. Their arrangement on

the three replicons [17] is di¡erent from that of aceA1 or aceA2 in R. metallidurans CH34, indicating that there are di¡erences in the organization of genes in these phylogenetically related bacteria. In R. eutropha HF39, it is obvious that only ICL1 is speci¢c to glyoxylate cycle since only R. eutropha HF39vaceA1 failed to grow on acetate (Fig. 2). Reduced growth of mutants R. eutropha HF39vaceA1 and HF39vaceB1 on propionate may indicate an involvement of the glyoxylate cycle for provision of oxaloacetate during the function of the 2-methylcitric acid cycle like in E. coli [18]. Many microorganisms possess two distinct classes of MS which are encoded by di¡erent genes. In E. coli, aceB encodes the A-form and glcB the G-form. The Table 4 Accumulation of PHAs by R. eutropha HF39 and the aceA1, aceA2, aceB1 or mdh mutants Strain

Poly(3HB) content (% CDW) Growth on gluconate

Growth on acetate

R. eutropha HF39 HF39vaceA1 HF39vaceA2 HF39vaceB1 VG12 VG12 (pBBRaceA1)

72.4 57.0 58.6 72.4 58.4 74.2

77.4 (2.1) ND 72.1 (4.5) 75.0 (3.4) ND ND

(3.2) (5.1) (3.5) (3.9) (4.5) (1.6)

Cultivation was done in MM containing 0.5% (w/v) of the sodium salt of the indicated carbon source at 30‡C and 120 rpm. After cell harvest in the stationary growth phase, poly(3HB) content was determined gas chromatographically in whole cells as described in Section 2. Data are means of triplicate determinations ( W S.D. in parentheses). ND: these mutants do not grow on acetate; therefore, poly(3HB) contents could not be analyzed.

FEMSLE 11240 30-10-03

70

Z.-X. Wang et al. / FEMS Microbiology Letters 228 (2003) 63^71

A-form is predominant in cells growing on acetate, while the G-form is mainly induced by glyoxylate [19]. Similarly, two di¡erent MS isoenzymes, Mls1p and Mls2p, were identi¢ed in S. cerevisiae. However, it is considered that only Mls1p contributes the activity both speci¢c and limited to the glyoxylate cycle [15]. In addition, Mls1p is sensitive to carbon catabolite repression but nearly insensitive to nitrogen catabolite repression [15]. Also D. radiodurans possesses two MSs, which are encoded by aceB1 and aceB2 and located on two di¡erent chromosomes [16]. In R. eutropha HF39 like in R. metallidurans CH34 (http:// genome.ornl.gov/microbial/reut/) only one gene encoding MS has been identi¢ed. The arrangement of aceB in the chromosome of R. eutropha HF39 is very similar to that of R. metallidurans CH34. It is well demonstrated that R. eutropha H16 contains two circular chromosomes with sizes of 4.1 and 2.9 Mb in addition to a 0.44-Mb conjugative megaplasmid pHG1 [17]. Since R. eutropha HF39vaceB1 could still grow on acetate (Fig. 2) and still expressed about 40% of the MS activity found in the parent strain (Table 3), this may indicate that R. eutropha HF39 contains a second MS. However, a second gene encoding a protein homologous to MS could not be identi¢ed in the available (yet incomplete) R. eutropha H16 genome sequence. Disruption of the glyoxylate pathway by mutation of aceA in P. putida in addition to reduced £ux through isocitrate dehydrogenase is predicted to increase the £ux into de novo-fatty acid biosynthesis, and hence increase the amount of PHAmcl accumulated in the cells [20]. In contrast, disruption of either aceA gene resulted in R. eutropha HF39 in a signi¢cant decrease of the poly(3HB) content of the cells from gluconate. Furthermore, overexpression of aceA1 restored poly(3HB) yield in the Tn5-induced mdh mutant VG12 of R. eutropha HF39 from gluconate. Combining our results with data obtained from metabolic £ux analysis for biosynthesis of poly(3HB) from various carbon sources [21] indicates that an intact TCA cycle and glyoxylate bypass are bene¢cial for cells to accumulate poly(3HB).

[2]

[3]

[4]

[5] [6]

[7]

[8] [9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

Acknowledgements [18]

One of the authors (Z.X.W.) gratefully received a fellowship from the Deutscher Akademischer Austauschdienst (DAAD). This study was partially carried out within the framework of the Competence Network Go«ttingen ‘Genome research on bacteria’ (Genomik) ¢nanced by the German Federal Ministry of Education and Research (BMBF).

[20]

References

[21]

[1] Kornberg, H.L. and Krebs, H.A. (1957) Synthesis of cell constituents

[19]

from C2 -units by a modi¢ed tricarboxylic acid cycle. Nature 179, 988^991. Steinbu«chel, A. and Schlegel, H.G. (1991) Genetics of poly(L-hydroxyalkanoic acid) synthesis in Alcaligenes eutrophus. Mol. Microbiol. 5, 535^542. Schlegel, H.G., Kaltwasser, H. and Gottschalk, G. (1961) Ein Submersverfahren zur Kultur wassersto¡oxidierender Bakterien : Wachstumsphysiologische Untersuchungen. Arch. Mikrobiol. 38, 209^222. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Marmur, J. (1961) A procedure for the isolation of deoxyribonucleic acids from microorganisms. J. Mol. Biol. 3, 208^218. Bra«mer, C.O. and Steinbu«chel, A. (2001) The methylcitric acid pathway in Ralstonia eu-tropha : new genes identi¢ed involved in propionate metabolism. Microbiology 147, 2203^2214. Grunstein, M. and Hogness, D.S. (1975) Colony hybridization : a method for the isolation of cloned DNAs that contain a speci¢c gene. Proc. Natl. Acad. Sci. USA 72, 3961^3965. Strauss, E.C., Kobori, J.A., Siu, G. and Hood, L.E. (1986) Speci¢cprimer-directed DNA sequencing. Anal. Biochem. 154, 353^360. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of proteins utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248^254. Timm, A., Byrom, D. and Steinbu«chel, A. (1990) Formation of blends of various poly(3-hydroxyalkanoic acids) by a recombinant strain of Pseudomonas oleovorans. Appl. Microbiol. Biotechnol. 33, 296^301. Bra«mer, C.O. and Steinbu«chel, A. (2002) The malate dehydrogenase of Ralstonia eutropha and functionality of the C3 /C4 metabolism in a Tn5-induced mdh mutant. FEMS Microbiol. Lett. 212, 159^164. Atomi, H., Ueda, M., Hikida, M., Hishida, T., Teranishi, Y. and Tanaka, A. (1990) Peroxisomal isocitrate lyase of the n-alkane-assimilating yeast Candida tropicalis: gene analysis and characterization. J. Biochem. 107, 262^266. Serrano, J.A. and Bonete, M.J. (2001) Sequencing, phylogenetic and transcriptional analysis of the glyoxylate bypass operon (ace) in the halophilic archaeon Haloferax volcanii. Biochim. Biophys. Acta 1520, 154^162. Salanoubat, M., Genin, S. and Artiguenave, F. et al. (2002) Genome sequence of the plant pathogen Ralstonia solanacearum. Nature 415, 497^502. Hartig, A., Simon, M.M., Schuster, T., Daugherty, J.R., Yoo, H.S. and Cooper, T.G. (1992) Di¡erentially regulated malate synthase genes participate in carbon and nitrogen metabolism of Saccharomyces cerevisiae. Nucleic Acids Res. 20, 5677^5686. White, O., Eisen, J.A. and Heidelberg, J.F. et al. (1999) Genome sequence of the radioresistant bacterium Deinococcus radiodurans R1. Science 286, 1571^1577. Schwartz, E. and Friedrich, B. (2001) A physical map of the megaplasmid pHG1, one of three genomic replicons in Ralstonia eutropha H16. FEMS Microbiol. Lett. 201, 213^219. Textor, S., Wendisch, V.F., De Graaf, A.A., Mu«ller, U., Linder, M.I., Linder, D. and Buckel, W. (1997) Propionate oxidation in Escherichia coli: evidence for operation of a methylcitrate cycle in bacteria. Arch. Microbiol. 168, 428^436. Molina, I., Pellicer, M.T., Badia, J., Aguilar, J. and Baldoma, L. (1994) Molecular characterization of Escherichia coli malate synthase G. Di¡erentiation with the malate synthase A isoenzyme. Eur. J. Biochem. 224, 541^548. Klinke, S., Dauner, M., Scott, G., Kessler, B. and Witholt, B. (2000) Inactivation of isocitrate lyase leads to increased production of medium-chain-length poly(3-hydroxyalkanoates) in Pseudomonas putida. Appl. Environ. Microbiol. 66, 909^913. Shi, H., Shiraishi, M. and Shimizu, K. (1997) Metabolic £ux analysis for biosynthesis of PHB in A. eutrophus from various carbon sources. J. Ferment. Bioeng. 84, 579^587.

FEMSLE 11240 30-10-03

Z.-X. Wang et al. / FEMS Microbiology Letters 228 (2003) 63^71 [22] Srivastava, S., Urban, M. and Friedrich, B. (1982) Mutagenesis of Alcaligenes eutrophus by insertion of the drug-resistance transposon Tn5. Arch. Microbiol. 131, 203^207. [23] Bullock, W.O., Fernandez, J.M. and Stuart, J.M. (1987) XL1-Blue: a high e⁄ciency plasmid transforming recA Escherichia coli strain with L-galactosidase selection. Biotechniques 5, 376^379. [24] Simon, R., Priefer, U. and Pu«hler, A. (1983) A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram-negative bacteria. Biotechnology 1, 781^791.

71

[25] Hohn, C. and Collins, J. (1980) A small cosmid for e⁄cient cloning of large DNA fragments. Gene 11, 291^298. [26] Overhage, J., Priefert, H., Rabenhorst, J. and Steinbu«chel, A. (1999) Biotransformation of eugenol to vanillin by a mutant of Pseudomonas sp. strain HR199 constructed by disruption of the vanillin dehydrogenase (vdh) gene. Appl. Microbiol. Biotechnol. 52, 820^828. [27] Kovach, M.E., Phillips, R.W., Elzer, P.H., Roop, R.M. and Peterson, K.M. (1994) pBBR1MCS : a broad-host-range cloning vector. Biotechniques 16, 800^802.

FEMSLE 11240 30-10-03