Regulation of phasin expression and polyhydroxyalkanoate (PHA) granule formation in Ralstonia eutropha H16

Markus Pötter1, Mohamed H. Madkour2, Frank Mayer2 and Alexander Steinbüchel1

Institut für Mikrobiologie der Westfälischen Wilhelms-Universität Münster, Corrensstrasse 3, D-48149 Münster, Germany1
Institut für Mikrobiologie und Genetik der Georg-August-Universität Göttingen, Grisebachstrasse 8, D-37077 Göttingen, Germany2

Author for correspondence: Alexander Steinbüchel. Tel: +49 251 8339821. Fax: +49 251 8338388. e-mail: steinbu{at}uni-muenster.de


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Regulation of expression of the phasin PhaP, which is the major protein at the surface of polyhydroxyalkanoate (PHA) granules in Ralstonia eutropha H16, was studied and analysed at the molecular level. The regulation of PhaP expression is achieved by an autoregulated repressor, which is encoded by phaR in R. eutropha. The occurrence of PhaR homologues and the organization of phaR genes was analysed in detail in 29 different bacteria. Three kinds of molecule to which PhaR binds were identified in cells of R. eutropha, as revealed by gel-mobility-shift assays, DNaseI footprinting, cell fractionation, immunoelectron microscopy studies employing anti-PhaR antibodies raised against purified N-terminal hexahistidine-tagged PhaR and in vitro binding studies employing artificial PHA granules. PhaR binds upstream of phaP at two sites comprising the transcriptional start site plus the -10 region and a region immediately upstream of the -35 region of the {sigma}70 promoter of phaP, where two imperfect 12 bp repeat sequences (GCAMMAAWTMMD) were identified on the sense and anti-sense strands. PhaR also binds 86 bp upstream of the phaR translational start codon, where the {sigma}54-dependent promoter was identified. PhaR also binds to the surface of PHA granules. In the cytoplasm of a phaR{Omega}Km mutant of R. eutropha H16, increased quantities of PhaP were detected and the cells formed by this strain were much smaller and had many more PHA granules present than the wild-type. These data support the following model for the regulation of phaP expression. Under cultivation conditions not permissive for PHA biosynthesis or in mutants defective in PHA biosynthesis, PhaR binds to the phaP promoter region and represses transcription of this gene. After the onset of PHA biosynthesis, under conditions that are permissive for the formation of nascent granules, PhaR binds to PHA granules and phaP is transcribed. At the later stages of PHA accumulation, PhaR no longer binds to the granules and the transcription of phaP is again repressed. In addition to this, phaR expression is subject to autoregulation. Excess PhaR that has not bound to the phaP upstream region or to PHA granules binds to the phaR upstream region, thereby repressing its own transcription.

Keywords: phaR, phaP regulation, repressor, inclusion bodies, autoregulation of phaR

Abbreviations: GARG, goat-anti-rabbit IgG–gold; His6, hexahistidine; MM, mineral salts medium; PHA, polyhydroxyalkanoate; PHB, polyhydroxybutyrate; poly(3HB), poly(3-hydroxybutyrate)


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Polyhydroxyalkanoates (PHAs) represent storage compounds for carbon and/or energy and occur as water-insoluble inclusions (PHA granules) in a large variety of prokaryotes. In most bacteria, PHAs are accumulated in the cytoplasm if a carbon source is provided in excess and if any other essential nutrient is limited (Anderson & Dawes, 1990 ). These biopolyesters reveal thermoplastic and/or elastomeric properties similar to those of petrochemically produced polypropylene (Steinbüchel, 1991 ; Müller & Seebach, 1993 ). Due to their biodegradability and origin from renewable resources, PHAs have attracted much interest, and many technical applications have been developed for them in the last few decades (Anderson & Dawes, 1990 ; Hocking & Marchessault, 1994 ). Since the discovery of poly(3-hydroxybutyrate) [poly(3HB)], in Bacillus megaterium over 70 years ago (Lemoigne, 1926 ), many different hydroxyalkanoic acids have been identified as constituents of PHAs, comprising different carbon chain lengths and also various substituents at different positions of the R-pendant groups attached to the backbone of the polymer (Steinbüchel & Valentin, 1995 ). One of the best studied micro-organisms with respect to PHA accumulation is the Gram-negative bacterium Ralstonia eutropha. It synthesizes PHAs that consist of hydroxyalkanoic acids with a short carbon chain length, with 3-hydroxybutyric acid as the predominant constituent. Other hydroxyalkanoic acids have also been identified as constituents of PHAs (e.g. Holmes et al., 1981 ; Kunioka et al., 1988 ; Doi et al., 1990 ). Just recently, the biosynthesis of 3-mercaptoalkanoic-acid-containing polymers by R. eutropha has also been demonstrated and, with reference to the different linkage type in the polymer backbone, these biopolymers were designated as polythioesters (Lütke-Eversloh et al., 2001a , b ).

R. eutropha is regarded as a model organism, along with Allochromatium vinosum, for studying short-carbon-chain-length PHA biosynthesis in bacteria (Steinbüchel & Schlegel, 1991 ). The genes responsible for PHA biosynthesis in R. eutropha have been cloned and characterized (Slater et al., 1988 ; Schubert et al., 1988 ; Peoples & Sinskey, 1989a , b ) and comprise a ß-ketothiolase (phaA), an acetoacetyl-CoA reductase (phaB) and the PHA synthase (phaC). ß-Ketothiolase (EC 2 . 3.1.9) catalyses the first step of poly(3HB) biosynthesis, i.e. the condensation of two molecules of acetyl-CoA to acetoacetyl-CoA. Acetoacetyl-CoA is then reduced to R-(-)-3-hydroxybutyryl-CoA by an acetoacetyl-CoA reductase (EC 1.1.1.36). In the last step of the reaction, R-(-)-3-hydroxybutyryl-CoA is polymerized to poly(3HB) by PHA synthase, the key enzyme of PHA biosynthesis. The resulting PHA granules are coated with a layer of phospholipids and proteins, with phasins as the predominant compound. Phasins are a class of proteins of between 14 and 28 kDa in size that form a layer at the surface of the hydrophobic core of PHA granules; they also influence the number and size of PHA granules (Wieczorek et al., 1995 ; Pieper-Fürst et al., 1995 ; Steinbüchel et al., 1995 ). The phasin of R. eutropha is encoded by phaP, and the formation of PhaP is dependent on PHA biosynthesis and accumulation (Wieczorek et al., 1995 ). In R. eutropha, the amount of phasin present in cells has been shown to parallel the amount of PHA present in cells (York et al., 2001a , b ).

Adjacent to the phaCAB gene cluster, an ORF (designated phaR) has been detected in R. eutropha (Slater et al., 1998 ). In this bacterium, PhaR seems to be involved in the regulation of phasin and PHA biosynthesis (York et al., 2002 ). Other genes homologous to phaR have been found in several other PHA-accumulating bacteria (Rehm & Steinbüchel, 1999 ), such as Sinorhizobium meliloti (Tombolini et al., 1995 ; Povolo & Casella, 2000 ). Recently, a phaR homologue (encoding PhaR) was identified in Paracoccus denitrificans which was capable of binding to the intergenic regions of phaCphaP and phaPphaR, and it was proposed that PhaR functions as a negative regulator of phasin synthesis (Maehara et al., 1999 , 2001 ). Whereas much knowledge of PhaR has been gained by studying P. denitrificans, less is known about PhaR in the model organism R. eutropha. York et al. (2002) suggested that PhaR promotes poly(3HB) synthesis in R. eutropha by regulating the expression of PhaP and by possibly regulating the expression of additional proteins. The aim of this study was to reveal the molecular mechanisms by which PhaR couples the synthesis of PhaP to the presence of poly(3HB) in R. eutropha. To achieve this, the phaR gene product of R. eutropha was analysed by electron microscopy investigations and DNA-binding experiments and by detailed analyses of a knock-out mutant. In combination with the results of a study that has been published recently (York et al., 2002 ), a comprehensive model for the regulation of PHA granule formation in R. eutropha is suggested and discussed in this work.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, plasmids and growth conditions.
The bacterial strains and plasmids used in this study are listed in Table 1. Cells of R. eutropha were grown at 30 °C in mineral salts medium (MM) supplemented with filter-sterilized sodium gluconate (Schlegel et al., 1961 ). Solid media contained 1·5 % (w/v) purified agar. Tetracycline and kanamycin were used at final concentrations of 25 and 160 µg ml-1, respectively. Cells of Escherichia coli were cultivated at 37 °C in Luria–Bertani (LB) medium (Sambrook et al., 1989 ). Growth was monitored photometrically at 600 nm or with a Klett–Summerson photometer (Manostat, catalogue no. 76-550-220) using filter no. 54 (520–580 nm).


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains and plasmids used in this study

 
Isolation and manipulation of DNA.
Chromosomal DNA of R. eutropha H16 was isolated by the method of Marmur (1961) . Plasmid DNA was isolated by the method of Birnboim & Doly (1979) . DNA restriction fragments were purified with the Nucleotrap Kit (Machery-Nagel), as described by the manufacturer. Restriction enzymes, ligases and other DNA-manipulating enzymes were used according to the manufacturers’ instructions.

Transfer of DNA.
Competent cells of E. coli were prepared and transformed by using the CaCl2 procedure described by Hanahan (1983) .

DNA sequencing.
This procedure was done by using the Sequi Therm EXCEL TM II Long Read Cycle Sequencing Kit (Epicentre Technologies), IRD800-labelled oligonucleotides (MWG-Biotech) and a Li-Cor 4000L (Li-Cor Biosciences) automated sequencer (MWG-Biotech).

PCR amplifications.
All PCR amplifications of DNA were carried out as described by Sambrook et al. (1989) , employing Pfx-DNA-polymerase (Gibco-BRL) and an Omnigene HBTR3CM DNA Thermal Cycler (Hybaid).

Inactivation of phaR in R. eutropha H16 by insertion of {Omega}Km.
To amplify phaR from the genomic DNA of R. eutropha H16, oligonucleotides phaR_up (5'-AAAACGATCGCAGCACCATGTTCGTGCAGC-3') and phaR_rp (5'-AAAACGATCGATTCATTGCCTACGCACTCG-3') were designed (PvuI restriction sites are shown in bold). The 987 bp PCR product was cloned into pSK Bluescript-, to create pSKphaR. This plasmid was then digested with StuI. The linearized plasmid (pSKphaR) was ligated with {Omega}Km that had been recovered from SmaI-digested pSKsym{Omega}Km DNA, whose construction has been described by Overhage et al. (1999) . E. coli XL-1 Blue was transformed with the ligation mixture and transformants harbouring the resulting hybrid pSKphaR{Omega}Km, conferring kanamycin resistance, were obtained. To exchange the functional phaR gene for the inactivated one, phaR{Omega}Km was cloned into the suicide vector pSUP202 (Simon et al., 1983a ). To achieve this, phaR{Omega}Km was isolated from pSKphaR{Omega}Km after its digestion with PvuI and the fragment was ligated with PvuI-digested pSUP202. E. coli S17-1 was transformed with the ligation mixture and transformants harbouring the hybrid pSUP202phaR{Omega}Km, conferring resistance to tetracycline and kanamycin, were obtained. Subsequently, pSUP202phaR{Omega}Km was transferred to R. eutropha H16 by conjugation. Homogenotes resulting from a double crossover exhibiting a kanamycin-resistant phenotype were selected and distinguished from heterogenotes resulting from only a single crossover and exhibiting a tetracycline- plus kanamycin-resistant phenotype on MM agar plates containing the respective antibiotics. The genotype of homogenotes was controlled by PCR and DNA sequencing.

Cloning of phaR and purification of recombinant hexahistidine(His6)-tagged PhaR from recombinant E. coli.
For the cloning of phaR into E. coli, PCR was done using phaR_His_5 (5' - AAAAAA CATATG CATCAC CAC CAC CAC CACATGGCCACGACCAAAAAAGGCGC-3') as the sense primer and phaR_His_3 (5'-AAAAAATCTAGACAGCGTGCGGGATATGCG-3') as the anti-sense primer. These primers were deduced from the upstream and downstream regions, respectively, of the phaR gene of R. eutropha H16 (Slater et al., 1998 ). The phaRHis6 PCR product obtained was purified and ligated into pMa/c5-915 (Table 1), which harbours the cI857ts gene encoding the temperature-sensitive {lambda} repressor. The recombinant His6–PhaR (N-terminal fusion) was purified from E. coli TOP10 harbouring pMa/c5-914phaRHis6. Protein purification under native and denaturing conditions was conducted with Ni-NTA Spin Columns (Qiagen), as described by the manufacturer.

PAGE and Western immunoblotting.
Protein samples were resuspended in gel loading buffer [0·6% (w/v) SDS, 1·25% (w/v) ß-mercaptoethanol, 0·25 mM EDTA, 10% (v/v) glycerol, 0·001% (w/v) bromophenol blue, 12·5 mM Tris/HCl; pH 6·8] and separated in SDS 12·5% (w/v) polyacrylamide gels, as described by Laemmli (1970) . Proteins were stained with Coomasie brilliant blue R-250 (Weber & Osborn, 1969 ) or with silver (Heukeshofen & Dernick, 1985 ). Immunological detection of the PhaR protein blotted from the SDS-polyacrylamide gel onto PVDF membranes was done exactly as described by Towbin et al. (1979) .

Preparation and purification of antibodies.
Approximately 850 µg of the PhaR–His6 protein was dissolved in 500 µl Tris/HCl (pH 7·0). This was then separated by SDS-PAGE and blotted onto a PVDF membrane. The membrane was sent to Eurogentec (Herstal, Belgium) for the production of polyclonal anti-PhaR antiserum in rabbits. The IgG fraction of the serum was purified on a protein A-Sepharose CL-4B affinity column (Hjelm et al., 1972 ). To obtain highly monospecific antibodies against PhaR, the antiserum was subjected to an affinity purification according to the method of Olmsted (1981) with the modifications of Pieper-Fürst et al. (1994) . Antibodies against PhaP were available from previous studies (Wieczorek et al., 1995 ).

Primer extension.
The oligonucleotide primer PE_phaR (5'-ATCGCTGTTTGCGCGCTGCTGCACC-3') was 5'-labelled with IRD800. Total RNA (1 µg) extracted from R. eutropha H16 cells, which were grown under the storage conditions described above, was purified by using the RNeasy Mini Kit (Qiagen). The extension reaction was carried out at 50 °C for 50 min in a total volume of 50 µl in the presence of 200 U of Superscript II reverse transcriptase, as described by the manufacturer (Invitrogen). After its digestion with RNase A for 30 min at 37 °C, the sample was precipitated with ethanol, dissolved in H2O and then analysed on a 6% (w/v) polyacrylamide sequencing gel containing 8 M urea, employing the Li-Cor 4000L automated DNA sequencer. A sequencing reaction using the same labelled primer was run alongside the sample to determine the size of the primer-extension product. The phaR PCR product described in the ‘gel-mobility-shift assay’ was used as template for the sequencing reaction.

Gel-mobility-shift assay.
To generate fragments comprising the upstream and downstream regions of phaP and phaR that could be employed in this assay, PCR was done using the genomic DNA of R. eutropha H16 and the appropriate primers. Amplification using primers USP_phaP_shift (5'-GCAATCGCGCATCGTTGAACACGCA-3') and DSP_phaP_shift (5'-GCCGAGGATCCTGTGCGCATCGGAG-3') with subsequent digestion of the PCR products with PstI and SmaI produced fragments of 126, 232, 518 and 775 bp. PCR products generated using primers USP_phaR_shift (5'-CTGCAGGCCGCCACG-3') and DSP_phaR_shift (5'-CCAGGATGATCTGCAGCAAGA-3') were digested with PstI to give fragments of 330 and 685 bp. These DNA fragments (1·5 µg) were mixed with purified PhaR–His6 (10–2000 ng) in binding buffer [1 mM EDTA, 10 mM Tris/HCl (pH 7·0), 80 mM NaCl, 10 mM ß-mercaptoethanol, 5% (w/v) glycerol] in a total volume of 20 µl. Incubation was carried out for 40 min at room temperature. After this incubation, 10xloading dye [250 mM Tris/HCl (pH 7·5), 0·2% (w/v) bromophenol blue, 0·2% (w/v) xylene cyanol, 40% (w/v) glycerol] was added to the samples. DNA–protein complexes were separated from unbound DNA fragments in 8% (w/v) native polyacrylamide gels, using 2xTris/borate/EDTA buffer (Sambrook et al., 1989 ). After electrophoresis, the gels were stained with ethidium bromide and the bands were visualized with a UV light.

DNaseI footprinting.
DNaseI footprinting was performed by using non-radioactive probes containing the IRD800 label in combination with a Li-Cor sequencer. For this purpose, DNA probes were prepared as follows. The PCR products described in the ‘gel-mobility-shift assay’ were used as template DNA in PCR reactions with an IRD800-labelled primer (footprint III, 5'-CCCGGAGTGGCGTCACAGCCGCTCCC-3') in combination with DSP_phaP_shift (5'-GCCGAGGATCCTGTGCGCATCGGAG-3'), and USP_phaR_shift (5'-CTGCAGGCCGCCACG-3') in combination with PE_phaR (5'-ATCGCTGTTTGCGCGCTGCTGCACC-3'). Ten nanograms of the IRD800-labelled fragment was used per reaction. The binding reaction conditions for DNaseI footprinting were identical to the conditions used in the ‘gel-mobility-shift assay’. DNaseI cleavage was done by adding 20 µl of a solution containing 5 mM CaCl2, 10 mM MgCl2 and 2·5 mU of DNaseI (Gibco-BRL) to the PCR products. After 1 min, the DNaseI reaction was stopped by the addition of 20 µl of 4 M ammonium acetate and 30 mM EDTA. The DNA was extracted with 60 µl of phenol, precipitated with 96% ethanol in the presence of 40 µl of 50% glycogen and washed with 70% (v/v) ethanol. The pellet was dissolved in 1 µl of formamide loading buffer, heated at 95 °C for 5 min and then chilled on ice. Subsequently, 0·8 µl of the sample was analysed on a Li-Cor 4000L sequencer using a 6% denaturating sequence gel with 0·2 mm spacers and settings at 2000 V, 25 mA, 50 W and 45 °C.

Electron microscopy studies.
Cells were washed and suspended in 50 mM potassium phosphate buffer (pH 6·8), fixed in the presence of a mixture of 0·2% (v/v) glutaraldehyde plus 0·3% (w/v) paraformaldehyde and embedded in Spurr’s low-viscosity resin (Spurr, 1969 ), as described by Walther-Mauruschat et al. (1977) . For the post-embedding and immunogold labelling of PhaR and PhaP, cells were embedded in Lowicryl K4M (Lowri) as described by Roth et al. (1981) , except that methanol was used instead of ethanol for dehydration. Immunological detection of PhaR and PhaP in ultra-thin sections, employing the primary antibodies and goat-anti-rabbit IgG–gold (GARG) complex (Dakopatts), was done exactly as described by Pieper-Fürst et al. (1994) . The specificity of the labelling was demonstrated by a control experiment using only the GARG complexes and mAbs against the Cap proteins of Sendai virus. Micrographs were taken with a Philips EM301 electron microscope at an acceleration voltage of 80 kV. Magnifications were calibrated with a cross-lined grating replica (Balzers).

Analysis of the PHA.
The samples [i.e. their poly(3HB)] were subjected to methanolysis in the presence of 15% (v/v) sulfuric acid. The resulting methyl esters of 3-hydroxybutyric acid were analysed by gas chromatography, as described by Brandl et al. (1988) and Timm & Steinbüchel (1990) .

Preparation of PHA granules.
Granules consisting of poly(3HB) were isolated from R. eutropha H16 cells that had been grown in sodium gluconate, by employing the hypochlorite treatment described previously (Jendrossek et al., 1993 ). PHA granules were also isolated by centrifugation in sucrose gradients according to the method of Preusting et al. (1993) with the modifications of Wieczorek et al. (1995) .


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Sequence analysis of the R. eutropha H16 phaR gene
Genes encoding putative proteins homologous to the R. eutropha phaR translational product are found in many PHA-accumulating bacteria (Fig. 1, and see Fig. I, supplementary data at http://mic.sgmjournals.org). However, the function of PhaR has not yet been revealed. The organization of phaR-like genes and of adjacent regions reported in the literature or as found by homology searches of microbial genomes are shown in Fig. II (supplementary data at http://mic.sgmjournals.org). Genes encoding PhaR homologues have, so far, only been found in the proteobacterial branch of the eubacteria (Fig. 1). An amino-acid-sequence alignment of PhaR homologues (Fig. I, supplementary data at http://mic.sgmjournals.org) revealed an overall high level of sequence identity in the N-terminal regions (amino acids 24–125), whereas the C-terminal regions (amino acids 126–225) lacked this striking similarity. Prediction of the secondary structure of the R. eutropha H16 PhaR sequence at the EMBL server (http://www.embl.de) revealed a putative helix–turn–helix motif in the region comprising amino acids 95–114 (Fig. I, supplementary data at http://mic.sgmjournals.org). With the exception of the PhaR homologue StdC, present in Comamonas testosteroni, all of the phaR homologues occurred within clusters of PHA biosynthesis genes (Fig. II, supplementary data at http://mic.sgmjournals.org). The location of these phaR homologues within the PHA biosynthesis gene clusters is variable; however, phaR is located close to phaA only in the genomes of the {alpha}-Proteobacteria Rhizobium etli, S. meliloti, Mesorhizobium loti, Caulobacter crescentus and Methylobacterium extorquens.



View larger version (32K):
[in this window]
[in a new window]
 
Fig. 1. Phylogenetic tree, constructed using the neighbour-joining method, of R. eutropha H16 and related bacteria based on a comparison of PhaR sequences. Accession numbers or references for each strain are given in parentheses (www.jgi.doe.gov, JGI Production Genomics Facility, California, USA; www.sanger-centre.ac.uk, The Sanger Centre, Cambridge, UK; www.tigr.org, The Institute for Genomic Research, Rockville, USA). Greek letters at the right of the tree indicate the subdivision of the Proteobacteria that the strains belong to. Bootstrap values for branches are given at the nodes. Bar, 0·1 aa substitutions.

 
Phenotypic characterization of the R. eutropha phaR{Omega}Km strain
Although a large number of transposon-induced mutants affected in PHA biosynthesis and accumulation have been screened and characterized at the molecular level (e.g. Schubert et al., 1988 ), so far no mutants of R. eutropha have been found that are defective in phaR. Therefore, no phenotype can be assigned to this gene. In this study, we generated a phaR knock-out mutant of R. eutropha by inserting a kanamycin-resistance cassette into phaR, employing a respective derivative of the suicide vector pSUP202 for homogenotization. When cells of the phaR{Omega}Km mutant were grown on MM containing 0·02% (w/v) NH4Cl (storage conditions) and sodium gluconate as sole carbon source, the cells accumulated poly(3HB) to 48% (w/w) of the cell dry weight in the stationary phase of growth, whereas the wild-type strain and the phaP polyhydroxybutyrate(PHB)-leaky mutant H2275 accumulated poly(3HB) to 76 or 31% (w/w) of the cell dry weight, respectively, if cultivated under the same conditions. In addition, cells of the phaR{Omega}Km mutants were clearly distinguishable from the wild-type and from the PHB-leaky mutant H2275 by the size and number of the poly(3HB) granules occurring in their cells. As shown in Fig. 2, cells of the phaR knock-out mutant contained a large number of small granules that filled almost the entire cytoplasm, whereas cells of the wild-type and of the PHB-leaky mutant H2275 contained fewer but larger granules or only one single, large granule, respectively.



View larger version (107K):
[in this window]
[in a new window]
 
Fig. 2. Morphology of cells of wild-type R. eutropha H16 (a) and of R. eutropha phaR{Omega}Km (b). Cells were cultivated in MM containing 0·02% (w/v) NH4Cl plus 1·5% (w/v) sodium gluconate and were harvested in the stationary phase of growth. Thin-sections were prepared and electron micrographs were obtained as described in Methods. Bars, 0·2 µm.

 
Expression of the N-terminal His6-tagged fusion protein of PhaR and its purification by metal-ion-chelating chromatography
Electropherograms of SDS-polyacrylamide gels in which crude extracts of the wild-type and a phaR{Omega}Km mutant had been separated did not reveal the presence or absence of a protein exhibiting an apparent molecular mass expected for the putative phaR translational product. Because PhaR probably has a regulatory function, only low concentrations of the protein were expected to be detected. Therefore, to generate a PCR product comprising the DNA regions directly upstream and downstream of the start and stop codons of phaR, respectively, tailed PCR was employed to amplify the coding region of phaR and to insert restriction sites for NdeI and XbaI and an 18 nt region encoding six His residues. This PCR product was inserted into the NdeI–XbaI sites of pMa/c-5-914. The resulting plasmid, pMa/cphaRHis6, contained phaR under the control of the temperature-sensitive {lambda}-repressor cI857ts, with an artificial ribosome-binding site conserved for E. coli. The His6–PhaR fusion protein could be detected in E. coli TOP10 crude extracts upon raising the temperature of the culture from 32 to 42 °C (Ausubel et al., 1987 ) and was purified from the recombinant cells of E. coli by applying the soluble fraction of crude extracts to a Ni-NTA/agarose column. The His6–PhaR fusion protein was eluted from the matrix with 250 mM imidazole; the imidazole was then removed by dialysis against 10 mM Tris/HCl (pH 7·0). The purity of phaR was analysed by SDS-PAGE (Fig. 3) and by immunoblotting (Fig. 4a, lane 5).



View larger version (77K):
[in this window]
[in a new window]
 
Fig. 3. Purification of PhaR. Following its expression in E. coli TOP 10, the His6-tagged PhaR fusion protein was purified using a Ni-NTA Spin Column under native conditions [lysis and wash in a solution containing 50 mM NaH2PO4, 300 mM NaCl and 20 mM imidazole (pH 8·0); elution with a solution containing 50 mM NaH2PO4, 300 mM NaCl and 250 mM imidazole (pH 8·0)]. Proteins were separated in 12·5% SDS-polyacrylamide gels and stained with Coomasie brilliant blue. Lanes: 1, crude protein extract from E. coli TOP10 pMa/c5-914; 2, crude protein extract from E. coli TOP10 pMa/cphaRHis6; 3, flow-through; 4, first wash step; 5, second wash step; 6–8, eluates; M, molecular mass marker.

 


View larger version (41K):
[in this window]
[in a new window]
 
Fig. 4. Localization of PhaR and binding of the PhaR His6-tagged protein of R. eutropha H16, overexpressed in E. coli TOP10 harbouring pMa/c5-914phaRHis6, to artificial poly(3HB) granules. (a) R. eutropha H16 was cultivated in MM containing 0·02% (w/v) NH4Cl plus 1·5% (w/v) sodium gluconate; cells were harvested in the stationary phase of growth. Crude protein extracts or proteins from solubilized granules were separated by SDS-PAGE and subjected to Western-blot analysis employing antibodies raised against PhaR. Lanes: 1, crude extract (300 µg total protein); 2, soluble fraction (55 µg); 3, membrane fraction (60 µg); 4, granule-associated proteins (32 µg); 5, purified PhaR His6-tagged protein overexpressed in E. coli TOP10 harbouring pMa/cphaRHis6 (15 µg). (b) Proteins were separated in 12·5% SDS-polyacrylamide gels and stained with silver. The purified PhaR His6-tagged protein (100 µg in 130 µl of 100 mM Tris/HCl, pH 7·0) was incubated for 90 min on ice with poly(3HB) granules. The mixtures were subsequently separated into pellets and supernatants by centrifugation and washed twice with 10 mM Tris/HCl, pH 7·5. Lanes: M, molecular mass marker; 1, artificial poly(3HB) granules incubated with the purified PhaR His6-tagged protein of E. coli TOP10 harbouring pMa/cphaRHis6; 2, supernatant after centrifugation; 3, supernatant after first wash step; 4, supernatant after second wash step; 5, pellet with PhaR His6-tagged protein after second wash step.

 
Specific binding of PhaR to the phaP and phaR upstream regions
To investigate the interaction of PhaR with DNA, gel-mobility-shift assays were performed (Fig. 5). Purified His6–PhaR was incubated with a mixture of fragments derived from PCR amplification and comprising phaP, phaR and their respective adjacent regions, as described in Methods. A 518 bp SmaI–PstI DNA fragment containing the phaP upstream and 5' phaP region clearly shifted, depending on the concentration of PhaR added. In contrast to this result, no retardations of the 128, 232 and 775 bp fragments were observed. (Fig. 5a). A 330 bp DNA fragment containing the phaR upstream and 5' phaR region clearly shifted at increasing concentrations of PhaR (Fig. 5b). None of the Sau3AI digestion products (i.e. the 119, 168 and 243 bp fragments) of the shifted 518 bp fragment shifted in presence of PhaR (data not shown). The gel-mobility-shift assays clearly revealed the binding of PhaR to phaP upstream regions, as well as to phaR upstream regions.



View larger version (38K):
[in this window]
[in a new window]
 
Fig. 5. Gel-mobility-shift assays of PhaR binding to DNA fragments containing the upstream regions of phaP (a) or phaR (b). (a) Different amounts of purified PhaR protein were incubated with 1·5 ng of DNA containing phaP and/or its upstream region. Lanes: 1, 0 ng; 2, 10 ng; 3, 100 ng; 4, 500 ng; 5, 1000 ng; 6, 2000 ng; M, 100 bp ladder. (b) Different amounts of purified PhaR protein were incubated with 1·5 ng of DNA containing phaR and/or its upstream region. Lanes: 1, 0 ng; 2, 100 ng; 3, 500 ng; 4, 2000 ng; M, 100 bp ladder. The DNA–protein complex was separated from unbound DNA on an 8% native polyacrylamide gel. DNA fragments were stained with ethidium bromide and visualized with UV light.

 
DNaseI footprinting experiments
To identify the PhaR-binding sites at the phaP and phaR upstream regions exactly, we performed DNaseI footprinting with purified His6–PhaR and PCR fragments containing the phaP or phaR promoter region (Fig. 6). The addition of PhaR to these samples resulted in DNaseI protection at a region of -49 to -36 and at a region of -12 to +21 relative to the phaP transcriptional start site (+1; Fig. 6a). Corresponding experiments were done to identify the exact binding site upstream of phaR. PhaR also mediated DNaseI protection at a region of +11 to +18 upstream of the phaR transcriptional start site (Fig. 6b).



View larger version (75K):
[in this window]
[in a new window]
 
Fig. 6. Footprinting analysis of PhaR binding. (a) DNaseI footprinting analysis using purified PhaR and phaP promoter DNA. Purified PhaR (0, 50, 2000 and 0 ng) was incubated with IRD800-labelled DNA. (b) Purified PhaR and phaR promoter DNA. Purified PhaR (0, 50 and 2000 ng) was incubated with IRD800-labelled DNA. After DNaseI treatment, the DNA was analysed in parallel with a sequencing reaction. The grey boxes indicate DNaseI protection, the open boxes refer to the sequence repeats and the arrows along the boxes indicate the direction of the repeats. Sequence positions are relative to the transcriptional start site (+1) of phaP or phaR.

 
Identification of the phaR promoter
Sequence analysis of phaR revealed a region 95 bp upstream of the translational start codon of phaR that exhibited extended homology with the eubacterial {sigma}54-promoter consensus sequence (Reitzer & Magasanik, 1985 ). The transcriptional start site of phaR was identified as being downstream of this region, by primer-extension assays with total RNA isolated from R. eutropha H16 that had been grown under storage conditions and harvested in the stationary phase of growth (Fig. 7). At 228 bp upstream of the translational start codon of phaR the nucleotide sequence TGT-N11-ACA was identified, which may represent a putative upstream activator sequence.



View larger version (54K):
[in this window]
[in a new window]
 
Fig. 7. Primer-extension assay of the transcript of phaR. Lanes A, C, G and T represent standard sequencing reactions to size the primer-extension product. A dot and an arrow indicate the origin (+1) and direction of transcription; the -12 and -24 regions are boxed.

 
Studies on the binding of PhaR to artificial poly(3HB) granules
Another putative location of PhaR is the PHA granule. Therefore, emulsions of artificial poly(3HB) granules and water (1·5%, w/v) were incubated with the purified His6–PhaR fusion protein for 90 min on ice (Wieczorek et al., 1995 ). After this incubation, the granules were collected by centrifugation, washed twice with 1 ml of 10 mM Tris/HCl (pH 7·0) and resuspended in denaturating buffer. The granule suspension and supernatants of each washing step were analysed by SDS-PAGE (Fig. 4b). The His6–PhaR fusion protein exhibited a high level of affinity for poly(3HB) (Fig. 4b, lane 5). Washing with 1 ml of 10 mM Tris/HCl (pH 7·0) did not remove the fusion protein from the poly(3HB) granules (Fig. 4b, lanes 2 and 3). To demonstrate that PhaR did not precipitate during the assay, PhaR was incubated without granules under the same conditions as described above. During the time course of this control experiment PhaR did not precipitate.

Localization of PhaR by Western immunoblotting
Because the in vitro binding studies shown above demonstrated the high affinity of His6–PhaR for artificial poly(3HB) granules, it was very likely that the native protein is one of the granule-associated proteins occurring in R. eutropha. Various methods were employed to study the in vivo localization of PhaR. The cellular localization of native PhaR was analysed by Western immunoblotting using monospecific polyclonal anti-PhaR antibodies to probe the native PHB granules fraction, the membrane fraction and the soluble protein fraction (see Methods for details of the fractionation procedure). In addition, the crude cellular extract was subjected to Western immunoblotting. Fig. 4(a) clearly shows that PhaR was detected in the crude extract (lane 1), in the insoluble fraction (lane 3) and in the poly(3HB) granule-associated protein fraction (lane 4), whereas no PhaR could be detected in the soluble protein fraction (lane 2). The experiments showed that the purified polyclonal antibodies against PhaR were highly specific for this protein. Only one band was obtained in Western blots prepared from crude extracts of E. coli TOP10 harbouring pMa/cphaRHis6 or of the poly(3HB) granule-associated protein fraction (Fig. 4a, lanes 5 and 4, respectively). To determine the quantities of the monospecific polyclonal anti-PhaR antibodies present, different concentrations (14, 28, 56, 112 and 224 µg) of the soluble protein fraction were subjected to Western immunoblotting and probing with the monospecific polyclonal anti-PhaR antibodies. The results clearly showed that PhaR was detected in the soluble protein fraction as a faint band only when 224 µg soluble protein was applied (data not shown).

Immunoelectron microscopic localization of PhaR
The monospecific polyclonal anti-PhaR antibodies were also used to localize PhaR in the cells at the ultrastructural level, by using immunoelectron microscopy. Cells of R. eutropha H16 and of the phaR knock-out mutant were embedded and ultra-thin sections were subjected to immunogold labelling. The labelling was mainly confined to only the periphery of the PHA granules in cells of the wild-type and there was no evidence that PhaR occurred within the core of the granules (Fig. 8). PhaR was not detected in the mutant.



View larger version (162K):
[in this window]
[in a new window]
 
Fig. 8. Immunoelectron microscopic detection and localization of PhaR in wild-type R. eutropha H16 (a) and R. eutropha phaR{Omega}Km (b). Cells were cultivated in MM containing 0·02% (w/v) NH4Cl plus 1·5% (w/v) sodium gluconate and were harvested in the stationary phase of growth. Thin-sections of cells were labelled with antibodies against the PhaR plus GARG complex. (a) R. eutropha H16 cell labelled with antibodies against the PhaR plus GARG complex. (b) Sections of cells of the phaR knock-out mutant R. eutropha phaR{Omega}Km labelled with antibodies against the PhaR plus GARG complex (control section). Bars, 0·2 µm.

 
Immunoelectron microscopic comparison of the R. eutropha H16 wild-type and R. eutropha phaR{Omega}Km cells with respect to the amount of PhaP present
Cells of the wild-type and of the phaR knock-out mutant were cultivated on MM containing sodium gluconate as carbon source under storage conditions and harvested in the stationary phase of growth. The cells were labelled with antibodies raised against PhaP. In the knock-out mutant, the amount of PhaP present (Fig. 9b), as represented by the GARG complexes, was clearly higher than the amount of PhaP present in the wild-type (Fig. 9a).



View larger version (132K):
[in this window]
[in a new window]
 
Fig. 9. Immunoelectron microscopic localization of PhaP in wild-type R. eutropha H16 (a) and the phaR knock-out mutant R. eutropha phaR{Omega}Km (b). Cells were cultivated as described in Fig. 8 and were labelled with antibodies against the PhaP plus GARG complex. Bars, 0·2 µm.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Previous studies have revealed that cells of R. eutropha are capable of synthesizing large amounts of PhaP, which contribute 3–5% of the total cellular protein (Wieczorek et al., 1995 ; Steinbüchel et al., 1996 ). However, the occurrence of PhaP in the cells is strictly dependent on PHA biosynthesis, and cells cultivated under conditions not permissive for PHA synthesis or mutants defective in the PHA synthase structural gene do not synthesize detectable levels of PhaP (Steinbüchel et al., 1995 ; Wieczorek et al., 1995 ; York et al., 2001a ). Moreover, even if PhaP is present in the wild-type in large quantities after employing permissive conditions, almost all of the PhaP is bound to the granules, and PhaP is hardly detected as a soluble protein in the cytoplasm (Wieczorek et al., 1995 ). To achieve all of the above, an effective mechanism for the regulation of PhaP formation is required, and this regulation most likely relies on at least one regulatory gene. Sequence analyses of the phaCAB operon of R. eutropha and the pha regions of many other bacteria have revealed the presence of a gene that exhibits substantial homology across bacterial species, which in R. eutropha H16 is referred to as phaR (Fig. I, supplementary data at http://mic.sgmjournals.org). PhaR and its homologous proteins from other bacteria have been predicted to be DNA-binding proteins (Cabrera et al., 1997 ; Povolo & Casella, 2000 ; Maehara et al., 2001 ).

As the conclusion of their physiological study, York et al. (2002) proposed a rather simple and straightforward mechanism for the regulatory action of PhaR in R. eutropha. According to these authors, PhaR is a repressor that binds to a regulatory sequence upstream of phaP; however, DNA-binding studies were not done by this group. Upon the onset of PHB synthesis, PhaR was postulated to be removed from the binding region, thus allowing the formation of PhaP; however, the ‘titrating factor’ was not identified. In the current study, we also found that PhaR binds upstream of phaP, as revealed by gel-mobility-shift assays (Fig. 5). Moreover, the exact location of the binding site was identified by DNaseI footprinting experiments. The latter experiments showed that PhaR protects two DNA regions: one region is located directly upstream of the -35 region of the promoter that was identified previously (Steinbüchel et al., 1996 ) and the second region maps from position -12 to +21. The protected regions comprise 14 and 33 bp and exhibit G+C ratios of 35 and 40 mol%, respectively. When the 14 and 33 bp binding sequences were analysed for sequence motifs that could serve as targets for DNA-binding proteins (e.g. palindromes, inverted or direct repeats), a 12 bp repeat sequence – GCAMMAAWTMMD – was identified on the sense and anti-sense strands, with the -10 and -35 regions of phaP located in between. These data clearly demonstrate (by the application of various independent methods) that PhaR binds to the promoter region of phaP. Maehara et al. (2001) have also shown in gel-mobility-shift assays that the PhaR protein of P. denitrificans specifically binds to the intergenic phaCphaP and phaPphaR regions.

This study also revealed that the PHA granules are a very plausible candidate for the ‘titrating factor’ of PhaR. The capability of PhaR to bind to PHA granules was demonstrated by in vitro experiments employing artificial granules as well as by in vivo cell fractionation and an immunoelectron microscopic approach employing specific anti-PhaR antibodies. Therefore, it seems plausible that upon the onset of PHA biosynthesis (i.e. when the nascent granules lack PhaP) PhaR is bound to the granules. Because of this, repression is removed from the phaP upstream binding region and phaP can be transcribed. The surface of the PHA granules is sufficiently large to bind the relatively small amounts of PhaR protein. However, the phasin PhaP most probably possesses a higher affinity for PHA granules than PhaR does. This could explain why in the later stages of PHA accumulation excess PhaR protein can again bind to the phaP upstream region, thus preventing the formation of more PhaP than can be bound to the granules.

By use of gel-mobility-shift assays, this study also clearly showed that PhaR binds not only to the phaP upstream region and to the PHA granules, but it also binds to the phaR upstream region. DNaseI footprinting experiments mapped the PhaR binding site to being approximately 86 bp upstream of the translational start site of phaR, which was also identified in this study. The transcriptional start site was identified as being 8 bp upstream of the putative -12 region. Normally, the distance from the transcriptional start to the -12 region is 4–8 bp (Hawley & McClure, 1983 ; Rosenberg & Court, 1979 ). Since the number of copies of the phaR upstream region is limited by the number of copies of chromosomes in the cell, this region has a lower capacity to titrate significant quantities of PhaR than the surface of the PHA granules. However, this binding may allow the efficient autoregulation of PhaR expression and, thus, may prevent the synthesis of more PhaR than is required for the repression of PhaP expression. Steinbüchel et al. (1996) have already suggested an autoregulation mechanism for the expression of PhaP; however, we have found that phaR (the repressor gene) is itself autoregulated and that phaP is not. Fig. 10 shows a model for the regulatory action of PhaR and its interactions with the phaP and phaR upstream regions and the PHA granules. This model is based on the results of this study and is also consistent with the results of previous studies (Wieczorek et al., 1995 ; Steinbüchel et al., 1996 ; Maehara et al., 2001 ; York et al., 2002 ).



View larger version (26K):
[in this window]
[in a new window]
 
Fig. 10. Suggested model for the regulation of PhaR and PhaP expression by PhaR in R. eutropha H16. The structural genes of phaR and phaP and their adjacent regions are shown. PhaR is shown as an oval structure, whereas PhaP is shown as a mushroom-like structure. Solid arrows, non-PHA accumulation conditions; open arrows, after the onset of PHA synthesis and its accumulation; dotted arrows, PHA accumulation has been finished; dashed arrows, translation of the respective protein (for further details see text).

 
The regulatory model presented in Fig. 10 is consistent with the phenotype of phaP and phaR mutants. phaP mutants do not synthesize phasin and, since the naked PHA granules exhibit a hydrophobic surface and are therefore not protected from coalescence, at the end of the accumulation phase the cells contain only one single, large granule. It has been shown previously that the introduction of multiple copies of phaP into R. eutropha results in the formation of many more, much smaller PHA granules than seen in the wild-type (Wieczorek et al., 1995 ). The phaR mutant generated in this study exhibited exactly this phenotype. This phenotype is explained by the lack of the repressor (PhaR) in this knock-out mutant, allowing the formation of increased amounts of PhaP; this explanation was confirmed by the immunoelectron microscopic studies performed here. Due to the higher concentration of PhaP in the cells, a larger surface area of PHA granules can be covered by PhaP, thus allowing stabilization of the PHA in water dispersion with a smaller size of granule than seen in the phaR mutant.

In conclusion, the interactions of the autoregulated repressor PhaR with the PHA granules and the phaP and phaR upstream promoter regions allow R. eutropha to regulate the expression of the phasin PhaP very efficiently. On the one hand, this regulation guarantees that the required large amounts (3–5% of total cellular protein) of PhaP can be synthesized if PHAs are synthesized. On the other hand, it ensures that PhaP is not synthesized in excess, thereby decreasing the burden imposed on the cells.


   ACKNOWLEDGEMENTS
 
This study was supported by a grant provided by the Deutsche Forschungsgemeinschaft (STE 386/6-1).


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Anderson, A. J. & Dawes, E. A. (1990). Occurrence, metabolism, metabolic role, and industrial uses of bacterial polyhydroxyalkanoates. Microbiol Rev 54, 450-472.

Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A. & Struhl, K. (1987). Current Protocols in Molecular Biology. New York: Wiley.

Birnboim, H. C. & Doly, J. (1979). A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res 277, 1513-1523.

Brandl, H., Gross, R. A., Lenz, R. W. & Fuller, R. C. (1988). Pseudomonas oleovorans as a source of poly(ß-hydroxyalkanoates) for potential applications as biodegradable polyesters. Appl Environ Microbiol 66, 2117-2124.[Abstract/Free Full Text]

Cabrera, J. E., Panzetta-Dutari, G., Pruneda, J. L. & Genti-Raimondi, S. (1997). A new Comamonas testosteroni steroid-inducible gene: cloning and sequence analysis. J Steroid Biochem Mol Biol 63, 91-98.[Medline]

Doi, Y., Segawa, A., Nakamura, S. & Kunioka, M. T. (1990). Production of biodegradable copolyesters by Alcaligenes eutrophus. In New Biosynthetic Biodegradable Polymers of Industrial Interest from Microorganisms , pp. 37-48. Edited by E. A. Dawes. Dordrecht:Kluwer.

Hanahan, D. (1983). Studies on transformation of Escherichia coli with plasmids. J Mol Biol 166, 557-580.[Medline]

Hawley, D. K. & McClure, W. R. (1983). Compilation and analysis of Escherichia coli promoter sequences. Nucleic Acids Res 11, 2234-2255.

Heukeshofen, J. & Dernick, R. (1985). A simplified method for silver staining of proteins in polyacrylamide gels and the mechanism of silver staining. Electrophoresis 6, 103-112.

Hjelm, H., Hjelm, K. & Sjöquist, J. (1972). Protein A from Staphylococcus aureus. Its isolation by affinity chromatography and its use as an immunosorbent for isolation of immunoglobulins. FEBS Lett 28, 73-76.[Medline]

Hocking, P. J. & Marchessault, R. H. (1994). Biopolyesters. In Chemistry and Technology of Biodegradable Polymers , pp. 48-96. Edited by G. J. L. Griffin. London:Chapman & Hall.

Holmes, P. A., Wright, L. F. & Collins, S. H. (1981). Betahydroxybutyrate polymers. European Patent Application 0052459.

Jendrossek, D., Knoke, I., Habidian, R. D., Steinbüchel, A. & Schlegel, H. G. (1993). Degradation of poly(3-hydroxybutyrate), PHB, by bacteria and purification of a novel PHB depolymerase from Comamonas sp. J Environ Polym Degrad 1, 53-63.

Kunioka, M., Nakamura, Y. & Doi, Y. (1988). New bacterial copolyesters produced in Alcaligenes eutrophus from organic acids. Polym Commun 29, 174-176.

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685.[Medline]

Lemoigne, M. (1926). Produits de deshydration et de polymerisation de lácide ß-oxybutyrique. Bull Soc Chim Biol 8, 770-782.

Lütke-Eversloh, T., Bergander, K., Luftmann, H. & Steinbüchel, A. (2001a). Identification of a new class of biopolymer: bacterial synthesis of a sulfur-containing polymer with thioester linkages. Microbiology 147, 11-19.[Abstract/Free Full Text]

Lütke-Eversloh, T., Bergander, K., Luftmann, H. & Steinbüchel, A. (2001b). Biosynthesis of poly(3-hydroxybutyrate-co-3-mercaptobutyrate) as a sulfur analogue to poly(3-hydroxybutyrate) (PHB). Biomacromolecules 2, 1061-1065.[Medline]

Maehara, A., Ueda, S., Nakano, H. & Yamane, T. (1999). Analyses of a polyhydroxyalkanoic acid granule-associated 16-kilodalton protein and its putative regulator in the pha locus of Paracoccus denitrificans. J Bacteriol 181, 2914-2921.[Abstract/Free Full Text]

Maehara, A., Doi, Y., Nishiyama, T., Takagi, Y., Ueda, S., Nakano, H. & Yamane, T. (2001). PhaR, a protein of unknown function conserved among short-chain-length polyhydroxyalkanoic acids producing bacteria, is a DNA-binding protein and represses Paracoccus denitrificans phaP expression in vitro. FEMS Microbiol Lett 200, 9-15.[Medline]

Marmur, J. (1961). A procedure for the isolation of deoxyribonucleic acids from microorganisms. J Mol Biol 3, 208-218.

Müller, H. M. & Seebach, D. (1993). Poly(hydroxyfettsäureester), eine fünfte Klasse von physiologisch bedeutsamen organischen Biopolymeren? Angew Chem 105, 483-509.

Olmsted, J. B. (1981). Affinity purification of antibodies from diazotized paper blots of heterogeneous protein samples. J Biol Chem 256, 11955-11957.[Abstract/Free Full Text]

Overhage, J., Priefert, H., Rabenhorst, J. & Steinbüchel, 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.[Medline]

Peoples, O. P. & Sinskey, A. J. (1989a). Poly-ß-hydroxybutyrate biosynthesis in Alcaligenes eutrophus H16. Characterization of the genes encoding ß-ketothiolase and acetoacetyl-CoA reductase. J Biol Chem 264, 15293-15297.[Abstract/Free Full Text]

Peoples, O. P. & Sinskey, A. J. (1989b). Poly-ß-hydroxybutyrate (PHB) biosynthesis in Alcaligenes eutrophus H16. Identification and characterization of the PHB polymerase gene (phbC). J Biol Chem 264, 15298-15303.[Abstract/Free Full Text]

Pieper-Fürst, U., Madkour, M. H., Mayer, F. & Steinbüchel, A. (1994). Purification and characterization of a 14-kilodalton protein that is bound to the surface of polyhydroxyalkanoic acid granules in Rhodococcus ruber. J Bacteriol 176, 4328-4337.[Abstract]

Pieper-Fürst, U., Madkour, M. H., Mayer, F. & Steinbüchel, A. (1995). Identification of the region of a 14-kilodalton protein of Rhodococcus ruber that is responsible for the binding of this phasin to polyhydroxyalkanoic acid granules. J Bacteriol 177, 2513-2523.[Abstract]

Povolo, S. & Casella, S. (2000). A critical role for aniA in energy–carbon flux and symbiotic nitrogen fixation in Sinorhizobium meliloti. Arch Microbiol 174, 42-49.[Medline]

Preusting, H., Kingma, J., Huisman, G., Steinbüchel, A. & Witholt, B. (1993). Formation of polyester blends by a recombinant strain of Pseudomonas oleovorans: different poly (3-hydroxyalkanoates) are stored in separate granules. J Environ Polym Degrad 1, 11-21.

Rehm, B. H. A. & Steinbüchel, A. (1999). Biochemical and genetic analysis of PHA synthases and other proteins required for PHA synthesis. Int J Biol Macromol 25, 3-19.[Medline]

Reitzer, L. J. & Magasanik, B. (1985). Expression of glnA in Escherichia coli is regulated at tandem promotors. Proc Natl Acad Sci USA 82, 1979-1983.[Abstract]

Rosenberg, M. & Court, D. (1979). Regulatory sequences involved in the promotion and termination of RNA transcription. Annu Rev Genet 13, 319-353.[Medline]

Roth, J., Bendayan, M., Carlemalm, E., Villiger, W. & Garavito, M. (1981). Enhancement of structural preservation and immunocytochemical straining in low temperature embedded pancreatic tissue. J Histochem Cytochem 29, 663-671.[Abstract]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Schlegel, H. G., Kaltwasser, H. & Gottschalk, G. (1961). Ein Submersverfahren zur Kultur wasserstoffoxidierender Bakterien: Wachstumsphysiologische Untersuchungen. Arch Mikrobiol 38, 209-222.

Schubert, P., Steinbüchel, A. & Schlegel, H. G. (1988). Cloning of the Alcaligenes eutrophus genes for synthesis of poly-ß-hydroxybutyric acid (PHB) and synthesis of PHB in Escherichia coli. J Bacteriol 170, 5837-5847.[Medline]

Simon, R., Priefer, U. & Pühler, A. (1983a). Vector plasmids for in vivo and in vitro manipulations of Gram negative bacteria. In Molecular Genetics of the Bacteria–Plant Interaction , pp. 98-106. Edited by A. Pühler. Berlin:Springer.

Simon, R., Priefer, U. & Pühler, A. (1983b). A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram-negative bacteria. Bio/Technology 1, 784-791.

Slater, S. C., Voige, W. H. & Dennis, D. E. (1988). Cloning and expression in Escherichia coli of the Alcaligenes eutrophus H16 poly-ß-hydroxybutyrate biosynthetic pathway. J Bacteriol 170, 4431-4436.[Medline]

Slater, S., Houmiel, K. L., Tran, M., Mitsky, T. A., Taylor, N. B., Padgette, S. R. & Gruys, K. J. (1998). Multiple ß-ketothiolases mediate poly(ß-hydroxyalkanoate) copolymer synthesis in Ralstonia eutropha. J Bacteriol 180, 1979-1987.[Abstract/Free Full Text]

Spurr, A. R. (1969). A low-viscosity epoxy resin embedding medium for electron microscopy. J Ultrastruct Res 26, 31-43.[Medline]

Steinbüchel, A. (1991). Polyhydroxyalkanoic acids. In Biomaterials , pp. 123-213. Edited by D. Byrom. Basingstoke, UK:Macmillan.

Steinbüchel, A. & Schlegel, H. G. (1991). Physiology and molecular genetics of poly(ß-hydroxy-alkanoic acid) synthesis in Alcaligenes eutrophus. Mol Microbiol 5, 535-542.[Medline]

Steinbüchel, A. & Valentin, H. E. (1995). Diversity of microbial polyhydroxyalkanoic acids. FEMS Microbiol Lett 128, 219-228.

Steinbüchel, A., Aerts, K., Babel, W. & 8 other authors (1995). Consideration on the structural and biochemistry of bacterial polyhydroxyalkanoic acid inclusions. Can J Microbiol 41, 94–105.[Medline]

Steinbüchel, A., Wieczorek, R. & Krüger, N. (1996). PHA biosynthesis, its regulation and application of C1-utilizing microorganisms for polyester production. In Microbial Growth on Compounds , pp. 237-244. Edited by M. E. Lidstrom & F. R. Tabita. Dordrecht:Kluwer.

Timm, A. & Steinbüchel, A. (1990). Formation of polyesters consisting of medium-chain-length 3-hydroxyalkanoic acids from gluconate by Pseudomonas aeruginosa and other fluorescent pseudomonads. Appl Environ Microbiol 56, 3360-3367.[Medline]

Tombolini, R., Povolo, S., Buson, A., Squartini, A. & Nuti, M. P. (1995). Poly-ß-hydroxybutyrate (PHB) biosynthetic genes in Rhizobium meliloti 41. Microbiology 141, 2553-2559.[Abstract]

Towbin, H., Staehelin, T. & Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 76, 4350-4354.[Abstract]

Walther-Mauruschat, A., Aragno, M., Mayer, F. & Schlegel, H. G. (1977). Micromorphology of Gram-negative hydrogen bacteria. II. Cell envelope, membranes, and cytoplasmic inclusions. Arch Microbiol 114, 101-110.[Medline]

Weber, K. & Osborn, M. (1969). The reliability of molecular weight determinations by dodecyl sulfate-polyacrylamide gel electrophoresis. J Biol Chem 244, 4406-4412.[Abstract/Free Full Text]

Wieczorek, R., Pries, A., Steinbüchel, A. & Mayer, F. (1995). Analysis of a 24-kilodalton protein associated with the polyhydroxyalkanoic acid granules in Alcaligenes eutrophus. J Bacteriol 177, 2425-2435.[Abstract]

York, G. M., Junker, B. H., Stubbe, J. A. & Sinskey, A. J. (2001a). Accumulation of the PhaP phasin of Ralstonia eutropha is dependent on production of polyhydroxybutyrate in cells. J Bacteriol 183, 4217-4226.[Abstract/Free Full Text]

York, G. M., Stubbe, J. & Sinskey, A. J. (2001b). New insight into the role of the PhaP phasin of Ralstonia eutropha in promoting synthesis of polyhydroxybutyrate. J Bacteriol 183, 2394-2397.[Abstract/Free Full Text]

York, G. M., Stubbe, J. & Sinskey, A. J. (2002). The Ralstonia eutropha PhaR protein couples synthesis of the PhaP phasin to the presence of polyhydroxybutyrate in cells and promotes polyhydroxybutyrate production. J Bacteriol 184, 59-66.[Abstract/Free Full Text]

Received 18 March 2002; revised 8 April 2002; accepted 29 April 2002.