Genetically Modified Strains of Ralstonia eutropha H16 with β-Ketothiolase Gene Deletions for Production of Copolyesters with Defined 3-Hydroxyvaleric Acid Contents

Abstract

β-Ketothiolases catalyze the first step of poly(3-hydroxybutyrate) [poly(3HB)] biosynthesis in bacteria by condensation of two acetyl coenzyme A (acetyl-CoA) molecules to acetoacetyl-CoA and also take part in the degradation of fatty acids. During growth on propionate or valerate, Ralstonia eutropha H16 produces the copolymer poly(3-hydroxybutyrate-co-3-hydroxyvalerate) [poly(3HB-co-3HV)]. In R. eutropha, 15 β-ketothiolase homologues exist. The synthesis of 3-hydroxybutyryl-CoA (3HB-CoA) could be significantly reduced in an 8-fold mutant (Lindenkamp et al., Appl. Environ. Microbiol. 76:5373–5382, 2010). In this study, a 9-fold mutant deficient in nine β-ketothiolase gene homologues (phaA, bktB, H16_A1713, H16_B1771, H16_A1528, H16_B0381, H16_B1369, H16_A0170, and pcaF) was generated. In order to examine the polyhydroxyalkanoate production capacity when short- or long-chain and even- or odd-chain-length fatty acids were provided as carbon sources, the growth and storage behavior of several mutants from the previous study and the newly generated 9-fold mutant were analyzed. Propionate, valerate, octanoate, undecanoic acid, or oleate was chosen as the sole carbon source. On octanoate, no significant differences in growth or storage behavior were observed between wild-type R. eutropha and the mutants. In contrast, during the growth on oleate of a multiple mutant lacking phaA, bktB, and H16_A0170, diminished poly(3HB) accumulation occurred. Surprisingly, the amount of accumulated poly(3HB) in the multiple mutants grown on gluconate differed; it was much lower than that on oleate. The β-ketothiolase activity toward acetoacetyl-CoA in H16ΔphaA and all the multiple mutants remained 10-fold lower than the activity of the wild type, regardless of which carbon source, oleate or gluconate, was employed. During growth on valerate as a sole carbon source, the 9-fold mutant accumulated almost a poly(3-hydroxyvalerate) [poly(3HV)] homopolyester with 99 mol% 3HV constituents.

INTRODUCTION

Polyhydroxyalkanoates (PHAs) are naturally occurring polyoxoesters that are synthesized and accumulated as cytoplasmic inclusions by diverse bacteria. Ralstonia eutropha strain H16, a Gram-negative facultatively chemolithoautotrophic hydrogen-oxidizing betaproteobacterium, accumulates poly(3-hydroxybutyrate) [poly(3HB)] in the form of insoluble granules as a storage compound for carbon and energy in the cytoplasm. The genome is composed of one megaplasmid and two chromosomes, whose nucleotide sequences were published in 2003 and 2006, respectively (26, 34). R. eutropha H16 harbors the PHA operon, which comprises three genes encoding a β-ketothiolase (phaA), an acetoacetyl coenzyme A (acetoacetyl-CoA) reductase (phaB), and a PHA synthase (phaC) (33). The β-ketothiolase (PhaA) condenses two acetyl-CoA molecules to acetoacetyl-CoA, and a stereospecific acetoacetyl-CoA reductase (PhaB) reduces the latter to R-(−)-3-hydroxybutyryl-CoA (24). Finally, the PHA synthase (PhaC) polymerizes the 3-hydroxybutyrate moieties of 3HB-CoA to poly(3HB). R. eutropha possesses two PHA synthases, of which only PhaC1 seems to contribute to the polymerization of the monomers (25, 33). PhaC1 belongs to the type I PHA synthases, which are known to produce short-chain-length PHAs (PHA_SCL) with 3 to 5 carbon atoms. However, during cultivation on fatty acids and in the presence of acrylate, which suppresses β-oxidation, this PHA synthase is also capable of incorporating small amounts of 3-hydroxyhexanoate (3HHx) and even 3-hydroxyoctanoate (3HO) into the polyester (7, 9). In addition to the enzymes mentioned above, granule-associated phasin proteins are crucial for the poly(3HB) metabolism in this bacterium (28).

More than 150 different PHA constituents besides 3HB are known to be constituents of microbial polyesters (40). In R. eutropha, poly(3HB) is produced predominantly as a homopolyester; however, this bacterium is capable of synthesizing copolyesters containing other hydroxyalkanoic acids also when fatty acids or other precursors are available or if the metabolism is altered by mutation (38). During heterotrophic growth on carbon sources such as gluconate, acetate, or fructose, the 3-hydroxyacyl-CoA thioesters are provided by de novo fatty acid biosynthesis (11, 13, 41). Carbohydrates are metabolized via the Entner-Doudoroff (2-keto-3-deoxy-6-phosphogluconate [KDPG]) pathway, yielding pyruvate, which is subsequently metabolized by oxidative decarboxylation to acetyl-CoA (12). Acetyl-CoA is then oxidized in the tricarboxylic acid cycle, used for anabolic pathways, or used for the synthesis of 3HA-CoA thioesters for PHA synthesis. In contrast, during growth on fatty acids, the 3HA-CoA thioesters are provided by β-oxidation (17).

The copolymer poly(3-hydroxybutyrate-co-3-hydroxyvalerate) [poly(3HB-co-3HV)] is accumulated by R. eutropha when adequate precursor substrates, e.g., propionic, levulinic, and valeric acid or propanol, are provided in the medium (6, 15, 19, 23). A necessary step in poly(3HB-co-3HV) biosynthesis in R. eutropha, which is catalyzed by the β-ketothiolase BktB, is the condensation of acetyl-CoA and propionyl-CoA to β-ketovaleryl-CoA (37).

Since the complete genome sequence of R. eutropha has become available, its metabolism can be exactly modified or engineered, and new features can be designed (5, 20). Since R. eutropha accumulates poly(3HB) to a level as high as 90% (wt/wt) of the dry weight of its cell, it is an interesting organism for large-scale production of PHAs and tailor-made polyester production by metabolic engineering (1, 18). Wild-type R. eutropha naturally accumulates copolymers that contain 3HB constituents. Thus, to obtain homopolymers other than poly(3HB), the production of 3HB-CoA has to be suppressed. Slater and coworkers postulated in 1998 that at least three β-ketothiolases are present in R. eutropha (37). However, the disclosure of the genome sequence in 2006 revealed in total 14 homologues in addition to phaA (26). In a recent study, several R. eutropha mutants lacking β-ketothiolases, which accumulate less poly(3HB) than the wild type, were generated (20). We obtained experimental evidence that the β-ketothiolase H16_A0170, besides PhaA and BktB, plays an important role in poly(3HB) synthesis in R. eutropha (20). In that previous study, we observed that mutants deficient in the three β-ketothiolases mentioned above, cultivated on sodium gluconate, accumulated poly(3HB) to only 30% (wt/wt) of the dry weight of the cell. Another homologue deleted at that time, H16_A1528, which showed no effect on poly(3HB) storage behavior, was recently identified as a component of one of the two β-oxidation operons present in R. eutropha (4). PcaF, a β-ketoadipyl CoA thiolase/acetyl-CoA acyltransferase, was still active in an 8-fold multiple mutant accumulating poly(3HB) to only 20% (wt/wt) of the dry weight of the cell, while genomewide transcriptome analyses of the wild type did not reveal this gene (20).

In the present study, we focused on investigation of the metabolic response of R. eutropha when different β-ketothiolases were lacking and on the residual capability to utilize fatty acids for growth and polymer synthesis. The deletion of pcaF resulted in the generation of a 9-fold mutant deficient in nine β-ketothiolases. We analyzed the PHA accumulation behavior of this new mutant, as well as those of various single and multiple β-ketothiolase mutants from the previous study, on propionate, valerate, octanoate, or oleate as the sole carbon source. None of the deleted β-ketothiolases affected the ability of R. eutropha to oxidize fatty acids during growth. BktB-deficient mutants synthesized almost a poly(3-hydroxyvalerate) homopolyester, poly(3HV), probably indicating a reduced 3HB-CoA pool. Moreover, the differences in β-ketothiolase enzyme activity and protein patterns between several R. eutropha β-ketothiolase mutants and the wild type, when cultivated on gluconate or oleate, were investigated.

MATERIALS AND METHODS

Bacterial strains and plasmids.

The bacterial strains and plasmids used in this study are listed in Tables 1 and 2, respectively.

Table 1.

Bacterial strains used in this study

Strain	Description	Reference/source
Ralstonia eutropha H16a
Wild type	Wild type	DSM 428
H16 ΔphaA	phaA precise-deletion gene replacement strain derived from R. eutropha H16	20
H16 ΔbktB	bktB precise-deletion gene replacement strain derived from R. eutropha H16	20
H16 ΔpcaF	pcaF precise-deletion gene replacement strain derived from R. eutropha H16	This study
H16 ΔH16_A0170	H16_A0170 precise-deletion gene replacement strain derived from R. eutropha H16	20
H16 ΔphaA ΔbktB (H16Δ2)	phaA and bktB deletion gene replacement strain derived from R. eutropha H16	20
H16 ΔphaA ΔbktB ΔH16_A0170 (H16Δ3)	phaA, bktB, and H16_A0170 deletion gene replacement strain derived from R. eutropha H16	20
H16 ΔphaA ΔbktB ΔH16_A1713 ΔH16_B1771 ΔH16_B1369 ΔH16_A1528 ΔH16_A0170 ΔH16_B0381 (H16Δ8)	phaA, bktB, H16_A1713, H16_B1771, H16_B1369, H16_A1528, H16_A0170, and H16_B0381 deletion gene replacement strain derived from R. eutropha H16	20
H16 ΔphaA ΔbktB ΔH16_A1713 ΔH16_B1771 ΔH16_B1369 ΔH16_A1528 ΔH16_A0170 ΔH16_B0381 ΔpcaF(H16Δ9)	phaA, bktB, H16_A1713, H16_B1771, H16_B1369, H16_A1528, H16_A0170, H16_B0381, and pcaF deletion gene replacement strain derived from R. eutropha H16	This study
Escherichia coli
Top10	F− mcrA Δ(mrr-hsdRMS-mcrBC)ϕ80lacZΔM15 ΔlacX74 deoR recA1 araD139 Δ(ara-leu)7697 galU galK rpsL endA1 nupG	Invitrogen
S17-1	thi proA hsdR17 hasdM+ recA; RP4-trafunction	36

^aDesignations for multiple mutants are given in parentheses.

Table 2.

Plasmids used in this study

Plasmid	Description	Reference/source
pJQ200mp18Tc	sacB oriV oriT traJ Tcr	29
pJQ200mp18Tc::ΔpcaF	ΔpcaF gene replacement plasmid; Tcr	This study

Media and growth conditions.

R. eutropha cells were cultivated aerobically at 30°C in 300-ml Erlenmeyer flasks with baffles containing 125 ml mineral salts medium (MSM) supplemented with 1% (wt/vol) sodium gluconate (Merck Schuchardt) or 0.1% to 0.3% (wt/vol) sodium valerate (Merck Schuchardt), sodium octanoate (Sigma-Aldrich), undecanoic acid (Sigma-Aldrich), or sodium oleate (Riedel-de Haën) as a carbon source (32). The detergent Triton X-100 at 1.5% (vol/vol) (Merck Schuchardt) was added to the medium to disperse undecanoic acid. The concentration of ammonium chloride was reduced to 0.05% (wt/vol) to provide conditions permissive for PHA accumulation. Escherichia coli cells were cultivated at 37°C in Luria-Bertani (LB) medium (31). Solid media contained 1.5% (wt/vol) agar agar. If required, tetracycline was added at the following concentrations: 12.5 μg ml⁻¹ for E. coli and 25 μg ml⁻¹ for R. eutropha. Growth of cells was measured photometrically at 600 nm with a spectrophotometer (Ultraspec 2000; Pharmacia Biotech). The addition of sodium oleate to the medium induces high turbidity; the growth of these cultures was therefore measured by determining the protein concentration by the method of Lowry et al. (21). All experiments were carried out in duplicate.

Isolation and manipulation of DNA.

Genomic DNA of R. eutropha was isolated by the method of Marmur (22). Plasmid DNA was isolated according to the protocol of Birnboim and Doly (2). DNA restriction fragments were purified from agarose gels by using the peqGOLD gel extraction kit (Peqlab) according to the manufacturer's instructions. Ligase and restriction endonucleases (Fermentas) were used according to the manufacturer's instructions.

Transfer of DNA.

Competent cells of E. coli were prepared and transformed with plasmids by the CaCl₂ procedure as described by Hanahan (10). Spot agar mating of R. eutropha or mutant derivatives with E. coli S17-1 as a plasmid donor was carried out on nutrient broth (NB) agar plates at 30°C. sacB gene selection was performed on NB agar plates supplemented with 10% (wt/vol) sucrose at 30°C.

PCR amplification.

DNA was amplified by PCR according to the method of Sambrook et al. by using Taq DNA polymerase (Invitrogen) in an Omnigene HBTR3CM DNA thermocycler (Hybaid) (31). The oligonucleotides employed for amplification are listed in Table 3.

Table 3.

Oligonucleotides used in this study

Oligonucleotide	Sequencea	Location
pcaF_XbaI_fw (flanking region 1)	AAATCTAGAATCCTGCTGCACAGCGAGAAC	5′ end of flanking region 1 (upstream of pcaF)
pcaF_EcoRI_rv (flanking region 1)	AAAGAATTCGGTGTGTCTCTCTCGTTTTGGTGTG	3′ end of flanking region 1 (upstream of pcaF)
pcaF_EcoRI_fw (flanking region 2)	AAAGAATTCGTCAAGCCAGTTGGAAAGCTGAG	5′ end of flanking region 2 (downstream of pcaF)
pcaF_XbaI_rv (flanking region 2)	AATCTAGACGGCGAAGCGCGAGGTCAC	3′ end of flanking region 2 (downstream of pcaF)
pcaF_fw (control)	AAAGACATTCCCGACGGTGCCGTG	5′ end of region upstream of pcaF
pcaF_rv (control)	TACCCATGCGCTTGACCGTGATGC	5′ end of region downstream of pcaF
pcaF_fw (internal control)	AAGACCAGGACCTGATGGCGCTG	5′ end of fragment within pcaF gene
pcaF_rv (internal control)	TTCGTTCAGCTCGATCACGTCGATC	3′ end of fragment within pcaF gene

^aRestriction sites are underlined.

DNA sequencing.

Sequencing reactions for DNA fragments were carried out according to standard procedures at the Sequence Laboratories Göttingen GmbH (Göttingen, Germany).

Generation of gene replacement strains employing the sacB system.

The flanking regions upstream (562 bp) and downstream (316 bp) of the target gene pcaF (H16_B0200) were amplified by PCR with the primer pairs pcaF_XbaI_fw/pcaF_EcoRI_rv and pcaF_EcoRI_fw/pcaF_XbaI_rv, respectively. The resulting fragments were digested with EcoRI and were ligated to yield an approximately 900 bp fragment. This fragment was then digested with XbaI and was cloned into the corresponding site of plasmid pJQ200mp18Tc. This precise-deletion gene replacement plasmid was then used to generate the corresponding single- or multiple-deletion mutant R. eutropha H16 ΔpcaF or R. eutropha H16Δ9 (see Table 1 for nomenclature). Deletion mutants were generated by adaptation of standard protocols by using plasmid pJQ200mp18Tc (27, 29). The donor strain E. coli S17-1 was then transformed with this suicide plasmid, and from there the plasmid was mobilized into the corresponding R. eutropha recipient strains (14). Mutants were identified on NB agar plates supplemented with 10% (wt/vol) sucrose and on mineral salts medium agar plates containing 25 μg ml⁻¹ tetracycline (27, 29). Correct gene replacement strains were confirmed by PCR analyses and DNA sequencing employing primers (pcaF_fw [control]/pcaF_rv [control]) that bind beyond the primers used for constructing the deletion gene replacement plasmids, as well as internal gene primers (pcaF_fw [internal control]/pcaF_rv [internal control]) for the detection of objectionable recombination events in other positions of the chromosome.

Analysis of PHA content.

R. eutropha cells were harvested by centrifugation (15 min, 3,500 × g, 4°C), washed in 0.9% (wt/vol) sodium chloride, and then lyophilized for 24 h. Cells grown on sodium oleate as a carbon source were washed with a mixture of water and hexane to remove the residual oleate. The PHA contents of the cells were determined upon methanolysis of 5 to 10 mg lyophilized cells in the presence of 85% (vol/vol) methanol and 15% (vol/vol) sulfuric acid. The resulting methyl esters of 3-hydroxybutyrate and 3-hydroxyvalerate were analyzed by gas chromatography (GC) as described previously (3, 41) by using an Agilent 6850 GC (Agilent Technologies, Waldbronn, Germany) provided with a BP21 capillary column (50 m by 0.22 mm; film thickness, 250 nm [SGE, Darmstadt, Germany]) and a flame ionization detector (Agilent Technologies).

Determination of β-ketothiolase activity.

For measurement of β-ketothiolase activity, cells of wild-type R. eutropha and of the mutants were grown in MSM containing 1% (wt/vol) sodium gluconate or 0.1% (wt/vol) sodium oleate as a carbon source. At the early-stationary phase, after 30 h of growth, cells were harvested by centrifugation (15 min, 3,500 × g, 4°C), washed twice with 0.9% (wt/vol) NaCl, and frozen until needed. Further, cells were resuspended in 10 mM Tris-HCl buffer containing 1 mM dithiothreitol (DTT), incubated for 10 min at 37°C with lysozyme, and disrupted by passing three times through a French press (Amicon, Silver Spring, MD). Cell debris was removed by centrifugation (15 min, 3,500 × g, 4°C), and the cell extracts obtained were used for the enzyme assay. The protein concentration was measured by the method of Lowry et al. (21). β-Ketothiolase activity was measured spectrophotometrically by the method described by Feigenbaum et al. in thiolysis reaction with slight modifications (8). The reaction mixture contained 10 mM Tris-HCl (pH 8.2), 25 mM MgCl₂, 33 μM acetoacetyl-CoA, and 90 μM CoA, and the reaction was started with the enzyme. The disappearance of Mg²⁺-enolate complexes with acetoacetyl-CoA was monitored at 30°C and 303 nm on a spectrophotometer (Evolution 160 UV-Vis; Thermo Fisher Scientific). The reaction progress was measured for 1 min. The extinction coefficient for acetoacetyl-CoA was 16.5 mM⁻¹ cm⁻¹. The mean value was estimated from two independent experiments. One unit of β-ketothiolase activity was defined as the amount of β-ketothiolase that catalyzes the cleavage of 1 μmol acetoacetyl-CoA to acetyl-CoA per min.

One-dimensional polyacrylamide gel electrophoresis (PAGE).

Cells of R. eutropha and its mutants were harvested, washed, and disrupted as described above. Cell debris was removed by centrifugation (15 min, 3,500 × g, 4°C), and the cell-free crude extracts obtained were used for gel electrophoresis. The protein concentration was measured by the method of Lowry et al. (21). Protein samples (20 μg protein) were resuspended in gel loading buffer (0.6% [wt/vol] sodium dodecyl sulfate [SDS], 1.25% [vol/vol] β-mercaptoethanol, 0.25 mM EDTA, 10% [vol/vol] glycerol, 0.001% [wt/vol] bromophenol blue, 12.5 mM Tris/HCl [pH 6.8]) and were separated in 12.5% (wt/vol) SDS-polyacrylamide gels as described by Laemmli (16). The proteins were stained with Coomassie brilliant blue R-250 (42).

RESULTS

Generation of mutants.

All mutants generated and used in this study are precise-deletion mutants; their nomenclature and detailed descriptions are shown in Table 1. Recently, a series of single and multiple deletion mutants was generated. This included an 8-fold mutant lacking eight β-ketothiolase genes: phaA, bktB, H16_A1528, H16_B1369, H16_A0170, H16_B0381, H16_A1713, and H16_B1771 (20). The deletion of only one β-ketothiolase gene did not result in significant differences in PHA accumulation or in growth behavior. In multiple mutants, the enzymes PhaA, BktB, and H16_A0170 showed the greatest effect on PHA storage and growth behavior. This triple mutant accumulated poly(3HB) only to 30% (wt/wt) of the dry weight of its cells; the 8-fold mutant accumulated poly(3HB) to only 20% (wt/wt) of the dry weight of its cells. Matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) analysis of a partially purified protein exhibiting β-ketothiolase activity (20) identified another homologue, H16_B0200 (PcaF), which was still active in the multiple mutants. For this reason, we decided to construct further deletion mutants in this study. A single mutant, H16 ΔpcaF, and a 9-fold mutant, H16Δ9, derived from the 8-fold mutant were generated by gene replacement employing the recombinant suicide vector pJQ200mp18Tc, which harbored the up- and downstream flanking regions of the target gene pcaF. This vector comprises the sacB system from Bacillus subtilis, which can be induced by the addition of sucrose to the medium and which is lethal when expressed in Gram-negative bacteria (29). By using this system, we generated markerless mutants without the introduction of antibiotic resistance markers.

Growth and PHA accumulation of β-ketothiolase mutants on different fatty acids.

During cultivation on sodium gluconate, the deletion of pcaF only did not lead to significant effects on growth and storage behavior compared to those of the wild type (data not shown). This was observed previously with the other single mutants (20). The multiple mutants H16Δ8 and H16Δ9 both accumulated poly(3HB) to approximately 15 to 20% (wt/wt) of the dry weight of the cells after 60 h of cultivation in MSM containing 1% (wt/vol) sodium gluconate.

Since we still do not know the functions of all the deleted β-ketothiolases, we examined the ability of mutants to store PHAs when different short- or long-chain and even- or odd-chain-length fatty acids were used as the sole carbon source. During growth on fatty acids, 3HB-CoA has to be generated via β-oxidation. Various mutants defective in different β-ketothiolases, and the wild type as a control, were cultivated under storage conditions permitting polymer accumulation in liquid MSM containing propionate, valerate, octanoate, undecanoic acid, or oleate as described in Materials and Methods. Samples were taken as indicated in Table 4 and the legends to Fig. 1 and 2 from each culture to analyze the polymer content.

Growth on even-chain-length fatty acids (C₈ and C₁₈).

Wild-type R. eutropha and the mutants H16Δ3, H16Δ8, and H16Δ9 were cultivated in MSM under storage conditions with an initial concentration of 0.1% (wt/vol) octanoate or oleate. To stimulate further growth, 0.1% (wt/vol) fatty acid was added to the culture after 13 h with octanoate and after 23 h with oleate.

No differences in the growth behavior of wild-type R. eutropha and the mutants on octanoate were observed. All strains reached the exponential-growth phase after 6 h, and after 60 h of cultivation, an optical density at 600 nm (OD₆₀₀) of about 6 was obtained (data not shown). Also, as shown in Table 4, no significant differences in PHA accumulation were noticed. All strains accumulated the homopolyester poly(3HB) to about 41.4 to 46.6% (wt/wt) of the dry weight of cells.

Table 4.

Biomass and PHA production of R. eutropha and multiple mutants cultivated in MSM containing valerate, octanoate, or oleate as the sole carbon source^a

Carbon source and strain	PHA content (% [wt/wt] of dry wt of cell)		PHA composition (mol%)				Dry wt (g liter−1) of cells after 60 h
	After 30 h	After 60 h	After 30 h		After 60 h
			3HB	3HV	3HB	3HV
Valerate
H16 (wild type)	36.2 ± 1.3	28.3 ± 1.2	42.5	57.5	39.7	60.3	1.2
H16Δ3	36.3 ± 0.7	28.2 ± 1.3	3.4	96.6	2.9	97.1	1.3
H16Δ8	36.3 ± 0.3	26.4 ± 2.2	2.8	97.2	2.8	97.2	1.3
H16Δ9	35.4 ± 1.1	30.0 ± 0.1	2.9	97.1	2.6	97.4	1.3
Octanoate
H16 (wild type)	46.6 ± 1.2	41.4 ± 0.4	100		100		1.2
H16Δ3	44.9 ± 0.7	42.7 ± 0.4	100		100		1.2
H16Δ8	45.8 ± 1.0	44.0 ± 2.5	100		100		1.2
H16Δ9	46.4 ± 0.7	41.4 ± 1.4	100		100		0.9
Oleate
H16(wild type)	60.8 ± 1.3	62.4 ± 5.0	100		100		3.1
H16Δ3	19.2 ± 1.2	38.7 ± 0.7	100		100		1.1
H16Δ8	18.4 ± 0.8	37.0 ± 0.8	100		100		1.0
H16Δ9	17.2 ± 2.3	38.4 ± 0.1	100		100		1.0

^aCells were cultivated under storage conditions in MSM with an initial concentration of 0.1% (wt/vol) valerate, octanoate, or oleate. Each culture was also enriched by subsequent addition of the same sole carbon source: 0.1% (wt/vol) valerate was added after 15 h and 23 h; 0.1% (wt/vol) octanoate was added after 13 h; and 0.1% (wt/vol) oleate was added after 23 h. The sodium salts of the organic acids were used.

Growth on oleate was determined by measuring the protein concentration by the method of Lowry et al. (21). The wild type reached a protein concentration of about 0.9 g liter⁻¹, whereas the mutant strains reached about 0.7 g protein liter⁻¹ after 60 h of cultivation. As shown in Table 4, all strains also synthesized a poly(3HB) homopolyester. The wild type had already accumulated poly(3HB) to approximately 60% (wt/wt) of the dry weight of the cells after 30 h; this value remained constant even after 60 h of cultivation. In contrast, all three mutant strains stored poly(3HB) at 17% to 19% (wt/wt) of the dry weight of the cells, and after 60 h, the polymer content doubled to approximately 38% (wt/wt) of the dry weight of the cells.

Growth on odd-chain-length fatty acids (C₃, C₅, and C₁₁).

Initially, as a first set of strains, wild-type R. eutropha and the multiple mutants H16Δ3, H16Δ8, and H16Δ9 were cultivated. Valerate (0.1% [wt/vol]) was used as a sole carbon source, and subsequent feeding of 0.1% valerate was done after 15 h and 23 h to stimulate further cell growth (Table 4). After 6 h, all strains reached the exponential-growth phase, and the OD₆₀₀s of the wild type and of all three mutant strains after 60 h were between 5 and 6 (data not shown). With valerate as a carbon source, all strains synthesized the copolymer poly(3HB-co-3HV), as expected. All strains accumulated similar amounts of copolymer: 36% (wt/wt) of the dry weight of cells after 30 h and approximately 30% (wt/wt) of the dry weight of cells after 60 h of cultivation time. The copolymers differed in their molar compositions of 3HB and 3HV: the wild-type copolymer consisted of about 40 mol% 3HB and 60 mol% 3HV, and the copolymers of all multiple mutants consisted of only about 2 to 3 mol% 3HB but 97 to 98 mol% 3HV (Table 4).

Since mutant H16Δ3, which lacked the β-ketothiolases phaA, bktB, and H16_A0170, accumulated the near-homopolymer poly(3HV), the single mutants lacking the homologues mentioned and a double mutant, H16Δ2, lacking phaA and bktB, were chosen as a second set of strains for cultivation on 0.2% (wt/vol) valerate for 75 h (Fig. 1). H16Δ3 and H16Δ9 were cultivated again as a control from the previous experiment. Mutant H16 ΔbktB reached the highest cell density (OD₆₀₀) of about 3.5 after 75 h. The double and triple mutants H16Δ2 and H16Δ3 showed similar growth behavior, with an OD₆₀₀ of 3 after 75 h. The H16Δ9 mutant yielded a higher OD₆₀₀ (2.5) than wild-type R. eutropha (OD₆₀₀, 2.25). The cultures of H16 ΔA0170 and H16 ΔphaA showed the lowest turbidity at 600 nm (OD₆₀₀, about 1.9) (Fig. 1A). The deletion of bktB yielded the largest amount of accumulated copolymer (30% [wt/wt] of the dry weight of cells), containing 95 mol% 3HV (Fig. 1B and C). The wild type and the single mutants H16 ΔphaA and H16 ΔA0170 accumulated comparable amounts of copolymer: 5 to 10% (wt/wt) of the dry weight of cells. The wild type and H16 ΔphaA incorporated similar amounts of 3HV (80 mol%) into the copolymer. Mutant H16 ΔA0170 exhibited the lowest fraction of 3HV constituents in the copolymer poly(3HB-co-3HV), with 60 mol%. The multiple mutants H16Δ2, H16Δ3, and H16Δ9 showed a copolymer content of 20 to 25% (wt/wt) of the dry weight of cells, with H16Δ9 cells reaching the highest fraction of 3HV (>99 mol%) (Fig. 1B and C). For comparison, undecanoic acid was chosen as another longer odd-chain-length fatty acid. It was added to the medium at 0.2% (wt/vol), and samples were withdrawn after 54 h. The copolymer of the H16Δ3 mutant again exhibited a higher fraction of 3HV (83.3 ± 0.6 mol%) than the wild type (20.8 ± 0.5 mol%). The H16Δ9 mutant still incorporated 10.8 ± 1.1 mol% 3HB into the copolymer while on valerate, and the near-homopolymer poly(3HV) was accumulated (data not shown). The single mutants H16 ΔphaA and H16 ΔA0170 showed no significant differences in polymer composition from the wild type (data not shown). Although the levels were not as high as on valerate as the sole carbon source, the single mutant H16 ΔbktB again reached higher fractions of 3HV (62.6 ± 2 mol%) than the wild type grown on undecanoic acid (data not shown).

Fig 1.

Growth behavior (A) and PHA accumulation (B and C) of R. eutropha H16 and various mutants in MSM containing 0.2% (wt/vol) valerate under storage conditions. PHA contents were analyzed by gas chromatography. (A) Growth of the wild type (+) and of the mutants H16 ΔA0170 (×), H16 ΔphaA (■), H16Δ3 (●), H16 ΔbktB (▲), H16Δ9 (◆), and H16Δ2 (*). (B) Poly(3HB-co-3HV) contents of cells. Bars for each strain, from left to right, follow the times at which samples were withdrawn for analysis (24 h, 36 h, 54 h, 75 h). Open portions of bars represent the 3HB contents, whereas shaded portions represent the 3HV contents (both expressed as percentages [weight/weight] of the dry weight of cells), in the copolymers. (C) Molar composition of the copolymer poly(3HB-co-3HV). Open bars represent the 3HB contents, and shaded bars represent the 3HV contents (both expressed as moles percent). Bars for each strain, from left to right, follow the times at which samples were withdrawn for analysis (24 h, 36 h, 54 h, 75 h).

All mutants deficient in bktB incorporated high fractions of 3HV into the copolymer. For that reason, the wild type and H16 ΔbktB were also chosen for cultivation on propionate in comparison to valerate as the sole carbon source (Fig. 2). The wild type on valerate as well as on propionate and H16 ΔbktB on propionate reached OD₆₀₀s of about 3.5. The mutant again reached a higher cell density (OD₆₀₀, 4.8) on valerate (data not shown). As shown in Fig. 2A and B, the deletion mutant H16 ΔbktB, grown on valerate, synthesized the largest amount of copolymer (36% [wt/wt] of the dry weight of cells), containing 95 mol% 3HV. On propionate, the copolymer composition switched nearly completely, to 96.5 mol% 3HB and 3.5 mol% 3HV (Fig. 2B).

Fig 2.

PHA accumulation by R. eutropha H16 and H16 ΔbktB in MSM containing 0.2% (wt/vol) propionate or 0.2% (wt/vol) valerate under storage conditions. After 30 h and 60 h, PHA contents were analyzed by gas chromatography. (A) Poly(3HB-co-3HV) contents of cells. Left bars show the polymer content after 30 h, and right bars show the polymer content after 60 h. Open portions of bars represent the 3HB contents, whereas shaded portions of bars represent the 3HV contents (both expressed as a percentage [weight/weight] of the dry weight of cells), in the copolymers. (B) Molar composition of the copolymer poly(3HB-co-3HV). Open portions of bars represent the 3HB contents, and shaded portions of bars represent the 3HV contents (both expressed as moles percent). C3, propionate; C5, valerate.

Comparison of β-ketothiolase activity in the wild type and the different mutant strains on gluconate and oleate.

Since the multiple mutants H16Δ3, H16Δ8, and H16Δ9 exhibited lower poly(3HB) synthesis behavior than the wild type, we decided to determine the β-ketothiolase activities of the wild type, the single mutants H16 ΔphaA, H16 ΔbktB, and H16 ΔA0170, and the multiple mutants H16Δ2, H16Δ3, and H16Δ9. For this purpose, the cells of these strains were cultivated on gluconate or oleate as the sole carbon source. Cells were harvested after 30 h of growth, and the crude extracts were analyzed for β-ketothiolase activity with an enzyme assay using acetoacetyl-CoA as the substrate. On gluconate as the sole carbon source, wild-type R. eutropha and the single mutants H16 ΔbktB and H16 ΔA0170 showed the highest β-ketothiolase activities, with 3.2, 3.6, and 3.7 U mg⁻¹, respectively (Table 5). The deletion of H16_A0170 or bktB alone had slightly increased the β-ketothiolase activity over that for the wild type. The deletion of phaA only exerted almost no effect on poly(3HB) storage behavior: the wild type and H16 ΔphaA accumulated poly(3HB) to around 80% (wt/wt) of the dry weight of cells during growth on gluconate. However, with 0.4 U mg⁻¹, a significant decrease in the β-ketothiolase activity toward acetoacetyl-CoA was determined for H16 ΔphaA during the enzyme assay. Surprisingly, the β-ketothiolase activities of the multiple mutants H16Δ2, H16Δ3, and H16Δ9 also remained at a level between 0.3 and 0.4 U mg⁻¹, though these strains exhibited a reduced ability to accumulate poly(3HB) on gluconate.

Table 5.

β-Ketothiolase activities of R. eutropha and multiple mutants cultivated under storage conditions in MSM supplemented with 1% (wt/vol) sodium gluconate or 0.1% (wt/vol) sodium oleate as the sole carbon source^a

Strain	Cultivation on gluconate		Cultivation on oleate
	Poly(3HB) content (% [wt/wt] of dry wt of cells)	β-Ketothiolase activity (U mg−1)	Poly(3HB) content (% [wt/wt] of dry wt of cells)	β-Ketothiolase activity (U mg−1)
H16 (wild type)	85 ± 1.1	3.2 ± 0.35	60 ± 1.3	3.3 ± 0.02
H16 ΔphaA	75 ± 1.4	0.4 ± 0.05	46 ± 2.1	0.4 ± 0.09
H16 ΔbktB	85 ± 0.6	3.6 ± 0.08	55 ± 4.4	4.0 ± 0.18
H16 ΔA0170	80 ± 2.8	3.7 ± 0.30	57 ± 0.5	4.8 ± 0.11
H16Δ2	50 ± 4.1	0.4 ± 0.10	25 ± 1.0	0.4 ± 0.10
H16Δ3	25 ± 0.7	0.3 ± 0.07	20 ± 0.9	0.3 ± 0.12
H16Δ9	15 ± 2.2	0.5 ± 0.29	17 ± 2.3	0.4 ± 0.19

^aAcetoacetyl-CoA served as the substrate. Sample preparation and enzyme assays were carried out as described in Materials and Methods.

With oleate as the sole carbon source, the relation of β-ketothiolase activities between the wild type and the mutant strains was similar to that with gluconate as the sole carbon source (Table 5). Wild-type R. eutropha and the single mutants H16 ΔbktB and H16 ΔA0170 exhibited the highest β-ketothiolase activities, with 3.3, 4.0, and 4.8 U mg⁻¹, respectively (Table 5). Thus, the absence of BktB or H16_A0170 again led to a slight increase in β-ketothiolase activity over that for the wild type. The deletion of phaA only resulted in a β-ketothiolase activity of 0.4 U mg⁻¹ and a slight decrease in the poly(3HB) content (Table 5). The β-ketothiolase activity of H16 ΔphaA was about 8 times lower than that of the wild type (3.3 U mg⁻¹) and 12 times lower than that of H16 ΔA0170 (4.8 U mg⁻¹). The thiolytic activities of the multiple mutants H16Δ2, H16Δ3, and H16Δ9 toward acetoacetyl-CoA remained at a level between 0.3 and 0.4 U mg⁻¹, though these strains showed reduced poly(3HB) contents in the cells.

The remaining protein samples were separated by 1-dimensional SDS-PAGE in order to obtain a rough overview of differences in protein expression (Fig. 3). The β-ketothiolases exhibit molecular masses between 40 and 43 kDa. The intensities of the protein bands about this size decreased clearly with each β-ketothiolase deletion, from the wild type to H16Δ9. MALDI-TOF analyses of a few selected bands were performed (Fig. 3). The circled band of the H16Δ9 sample (around 43 kDa) from the culture grown on oleate turned out to be the β-ketothiolase H16_A0462. The band of the same size of the culture grown on gluconate did not reveal any β-ketothiolase. Another band, around 60 kDa, which seemed to be expressed at higher levels in oleate cultures, attracted attention (Fig. 3). MALDI-TOF analysis revealed this band to be a mixture of H16_A0460 and an isocitrate lyase (H16_A2211). By a microarray study, Budde and colleagues were able to show that the operon comprising H16_A0459 to H16_A0464 in wild-type R. eutropha is involved in β-oxidation during growth on trioleate as a carbon source (4). Moreover, the isocitrate lyase H16_A2211 was shown to be upregulated under the same growth conditions (4). We supported these findings, since we could identify H16_A0460, H16_A0462, and H16_A2211 by the MALDI-TOF analysis of H16Δ9 grown on oleate as a sole carbon source.

Fig 3.

Soluble fractions of disrupted cells of R. eutropha and mutants grown in MSM supplemented in the presence of 1% (wt/vol) sodium gluconate (G) or 0.1% (wt/vol) sodium oleate (C₁₈) under storage conditions. Protein concentrations were measured by the method of Lowry et al. (21), and 20 μg of protein was applied to each lane in the gel. Proteins were separated in a 12.5% (wt/vol) SDS-polyacrylamide gel and were stained with Coomassie brilliant blue. The molecular mass range for the β-ketothiolases is boxed. Circled bands were analyzed by MALDI-TOF analysis.

DISCUSSION

Since the disclosure of the genome sequence of R. eutropha, a number of diverse precise gene deletion mutants have been generated by different laboratories (4, 5, 20, 25). The mutants obtained show a variety of features that might point to interesting characteristics for biotechnological applications. It is known that 15 β-ketothiolase homologues exist in R. eutropha (30). In this study, we examined several mutants, including a newly established 9-fold mutant, deficient in nine β-ketothiolases (phaA, bktB, H16_A1713, H16_B1771, H16_B1369, H16_A1528, H16_A0170, H16_B0381, and pcaF). We analyzed the growth and storage behavior of mutants deficient in different β-ketothiolases on diverse short- and long-chain, as well as odd- and even-chain-length, fatty acids in order to gather more information about the function of β-ketothiolases in fatty acid metabolism and to possibly modify the polymer composition in R. eutropha.

Comparison of the growth of the wild type and the multiple mutants on octanoate, a medium-chain fatty acid, failed to demonstrate any differences, and the cells of all strains reached similar dry weights of around 1 g liter⁻¹ and comparable poly(3HB) contents of about 41 to 46% (wt/wt) of the dry weight of their own cells (Table 4). Cultures grown on oleate differed in growth and storage behavior. In the wild type, poly(3HB) reached 60% (wt/wt) of the dry weight of cells after 30 h, whereas the multiple mutants H16Δ3, H16Δ8, and H16Δ9 synthesized poly(3HB) only at 17% to 19% (wt/wt) of the dry weight of cells after 30 h, a level that was doubled after 60 h of cultivation time. This indicates that the lack of PhaA, BktB, and H16_A0170 seemed to decelerate polymer biosynthesis on this long-chain fatty acid. During growth on gluconate under storage conditions, the multiple mutants H16Δ3 and H16Δ9 were able to accumulate poly(3HB) at only 25% (wt/wt) or 15% (wt/wt) of the dry weight of cells, respectively (Table 5). We carried out an enzyme assay of the crude extracts of the wild type, the single mutants H16 ΔphaA, H16 ΔbktB, and H16 ΔA0170, and the multiple mutants H16Δ2, H16Δ3, and H16Δ9 with acetoacetyl-CoA as the substrate. It could be demonstrated that activity with acetoacetyl-CoA was almost 10-fold lower in all mutants deficient in phaA (Table 5). The ability to accumulate poly(3HB) during growth on gluconate was not affected among the single mutants, but it decreased in the multiple mutants. Interestingly, β-ketothiolase activity with acetoacetyl-CoA remained almost constant for H16 ΔphaA and the multiple mutants. On the other hand, the multiple mutants accumulated larger amounts of poly(3HB) on oleate, when 3HB-CoA must be provided by β-oxidation, than on gluconate.

On valerate as the sole carbon source, we were almost able to produce the homopolymer poly(3HV), with >99 mol% 3HV in the polyester accumulated by H16Δ9. The poly(3HV) homopolymer is synthesized only by a very few microorganisms: for example, a recombinant Aeromonas hydrophila 4AK4 strain during cultivation on undecanoic acid and Chromobacterium violaceum during cultivation on valeric acid (35, 39). R. eutropha produces the copolymer poly(3HB-co-3HV) when odd-numbered-chain-length carbon sources such as propionate, valerate, or propanol are provided in the culture medium (23, 37). Here we chose valeric and undecanoic acids to analyze the influence of several β-ketothiolase deletions on the ability to produce poly(3HB-co-3HV) or the homopolymer poly(3HV). PhaA appears to be principally restricted to the biosynthesis of acetoacetyl-CoA, while BktB is able to complement the lack of PhaA and also catalyzes the synthesis of acetoacetyl-CoA and the synthesis of β-ketovaleryl-CoA during growth on fructose and propionate (37). Slater et al. demonstrated that BktB is able to complement PhaA for PHA production in E. coli (37). This is also true in R. eutropha regarding poly(3HB) accumulation. BktB-deficient cells lack the ability to mediate the condensation of propionyl-CoA with acetyl-CoA to β-ketovaleryl-CoA during growth on propionate (37). We could also observe this effect in H16 ΔbktB during cultivation on propionate (Fig. 2). On valerate, on the other hand, this mutant accumulated the copolymer poly(3HB-co-3HV) to 30% (wt/wt) of the dry weight of cells, with 96.4 mol% 3HV. Under the same conditions, the mutant H16Δ9 accumulated the near-homopolymer poly(3HV) containing >99 mol% 3HV (Fig. 1). The wild type, in contrast, accumulated the copolymer, which contained 80 mol% 3HV, only at 10% (wt/wt) of the dry weight of cells. Due to the fact that BktB is responsible for the condensation of propionyl-CoA with acetyl-CoA to form β-ketovaleryl-CoA, the valerate in mutants lacking bktB must be completely converted directly via valeryl-CoA to β-hydroxyvaleryl-CoA, which is then incorporated into the polymer.

In this study, we could demonstrate that when PhaA was absent in the crude extract, the remaining β-ketothiolase activity toward acetoacetyl-CoA in the mutant was almost 10-fold lower than that for the wild type. The capability of H16 ΔphaA for poly(3HB) biosynthesis was only slightly affected. Surprisingly, the β-ketothiolase activity in H16Δ9 remained like that in the single mutant H16 ΔphaA. Since the poly(3HB) content in the multiple mutant was much lower, it might be expected that the β-ketothiolase activity toward acetoacetyl-CoA would also decrease (Table 5). A potential candidate for this residual activity in the proteome was identified by MALDI-TOF analysis as H16_A0462 (Fig. 3). In a microarray study, Brigham and colleagues identified two fatty acid β-oxidation operons in R. eutropha, each comprising five genes that are upregulated during growth on trioleate (4). A double deletion mutant defective in both operons was not able to use palm oil or crude palm kernel oil as a sole carbon source (4). The β-ketothiolases H16_A0462 and H16_A1528 are components of these operons. Therefore, we suppose that these two proteins prefer long-chain fatty acid derivatives and show less or even no affinity toward acetoacetyl-CoA. It is also assumed that one or even several of the five remaining β-ketothiolases (H16_B0668, H16_A1720, H16_B0759, H16_B0662, H16_A1887) must contribute to the remaining poly(3HB) content.

In this study, we showed that the absence of nine β-ketothiolases did not affect the ability to utilize short-, medium-, or long-chain fatty acids as sole carbon sources, but it did have an effect on poly(3HB) synthesis. The remaining six β-ketothiolases in H16Δ9 were sufficient for the utilization of these carbon sources. By controlling the 3HB-CoA content, we could increase the molar fraction of 3HV. The mutants obtained can be also used for further studies, unraveling the regulation of fatty acid metabolism and consequent PHA synthesis, and to promote the production of tailor-made biopolyesters by metabolic engineering in R. eutropha.