Contents lists available at SciVerse ScienceDirect
Microbiological Research
j our na l ho mepage: www.elsev ier .de /micres
olecular characterisation of phaCAB from Comamonas sp. EB172 for functionalxpression in Escherichia coli JM109
ian-Ngit Yeea, Jo-Ann Chuahb, Mei-Ling Chongc, Lai-Yee Phanga, Abdul Rahim Rahaa, Kumar Sudeshb,ohd Ali Hassana,d,∗
Faculty of Biotechnology and Biomolecular Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, MalaysiaEcobiomaterial Research Laboratory, School of Biological Sciences, Universiti Sains Malaysia, 11800 Penang, MalaysiaACGT Sdn Bhd, Lot L3-l-1, Enterprise 4, Technology Park Malaysia, Bukit Jalil, 57000 Kuala Lumpur, MalaysiaFaculty of Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia
r t i c l e i n f o
rticle history:eceived 11 October 2011eceived in revised form2 December 2011ccepted 28 December 2011
eywords:olyhydroxyalkanoatehaCABCo operon
a b s t r a c t
In this study, PHA biosynthesis operon of Comamonas sp. EB172, an acid-tolerant strain, consisting ofthree genes encoding acetyl-CoA acetyltransferase (phaACo gene, 1182 bp), acetoacetyl-CoA reductase(phaBCo gene, 738 bp) and PHA synthase, class I (phaCCo gene, 1694 bp) were identified. Sequence analysisof the phaACo, phaBCo and phaCCo genes revealed that they shared more than 85%, 89% and 69% iden-tity, respectively, with orthologues from Delftia acidovorans SPH-1 and Acidovorax ebreus TPSY. The PHAbiosynthesis genes (phaCCo and phaABCo) were successfully cloned in a heterologous host, Escherichia coliJM109. E. coli JM109 transformants harbouring pGEM′-phaCCoABRe and pGEM′-phaCReABCo were shownto be functionally active synthesising 33 wt.% and 17 wt.% of poly(3-hydroxybutyrate) [P(3HB)]. E. coli
omamonas sp. EB172oly(3-hydroxybutyrate)
JM109 transformant harbouring the three genes from the acid-tolerant Comamonas sp. EB172 (phaCABCo)under the control of native promoter from Cupriavidus necator, in vivo polymerised P(3HB) when fedwith glucose and volatile mixed organic acids (acetic acid:propionic acid:n-butyric acid) in ration of3:1:1, respectively. The E. coli JM109 transformant harbouring phaCABCo could accumulate P(3HB) at2 g/L of propionic acid. P(3HB) contents of 40.9% and 43.6% were achieved by using 1% of glucose andmixed organic acids, respectively.
ntroduction
Polyhydroxyalkanoates (PHA) are biodegradable, water-nsoluble polyesters that are accumulated intracellularly asarbon storage compounds in the cytoplasm (Steinbüchel andüchtenbusch 1998; Li et al. 2009). PHA can be synthesised byumerous prokaryotes in response to stress conditions such as
imitations of nitrogen, phosphate, oxygen and other elementsssential for growth (Anderson and Dawes 1990). PHA has plastic-ike properties and has therefore attracted much attention as theext generation bio-based and biodegradable plastics (Sudeshnd Iwata 2008). In addition, the biocompatibility and non-toxicroperties of PHA have created niche applications in the medical,
ackaging, agriculture and cosmetics fields as substitutes for petro-hemical synthetic plastics (Braunegg et al. 1998; Sudesh et al.007; Bhubalan et al. 2011). However, the high PHA production
∗ Corresponding author at: Faculty of Biotechnology and Biomolecular Sciences,niversiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia. Tel.: +60 3 89467590;
cost is mainly from fermentation and recovery processes whichhave hindered the commercialisation process of PHA. In orderto overcome the high production cost, researchers have exten-sively investigated the usage of recombinant microbial strains,mixed cultures, cheap carbon substrates, efficient fermentationand recovery/purification process (Khanna and Srivastava 2005;Verlinden et al. 2007; Li et al. 2007).
PHA biosynthesis in bacteria mainly involves three basicenzymatic steps (Naik et al. 2008). These enzymatic steps areinvolved in the generation of PHA monomer units that are even-tually polymerised by PHA synthase. The genes responsible forthe biosynthesis of PHA are those encoding the PHA synthase(PhaC), acetyl-CoA acetyltransferase (PhaA) (also referred to as�-ketothiolase or simply thiolase) and acetoacetyl-CoA reduc-tase (PhaB) enzymes, which may be clustered in a single operon.The enzymes responsible for the monomer supply are thio-lase and reductase (Steinbüchel and Lütke-Eversloh 2003; Rehm2003). The pathway for the synthesis of P(3HB) begins with
the condensation of two molecules of acetyl-CoA to acetoacetyl-CoA by �-ketothiolase, which is encoded by the phaA gene.Acetoacetyl-CoA reductase, a product of the phaB gene, reducesthe acetoacetyl-CoA to (R)-3-hydroxybutyryl-CoA, and finally PHA
from C. necatorMatsusaki et al. (2000)and Bhubalan et al.(2011)
pGEM′-phaCCoABRe pGEM-T derivative, phaCCo
from Comamonas sp. EB172DSM 23953, phaABRe from C.necator
This study
pGEM′-phaCReABCo pGEM-T derivative, phaABCo
from Comamonas sp. EB172DSM 23953, phaCRe from C.necator
This study
pGEM′-phaCABCo pGEM-T derivative, phaCABCo
from Comamonas sp. EB172This study
cols. The amplified full-length genes were cloned into a pGEM-Teasy vector (Promega, Madison, WI, USA) followed by transforma-
L.-N. Yee et al. / Microbiolog
ynthase, encoded by the phaC gene, catalyses the polymerisationf (R)-3-hydroxybutyryl-CoA to form P(3HB). P(3HB) is an abun-antly occurring short-chain length PHA that constitutes a carboneserve in a wide variety of bacteria (Steinbüchel and Füchtenbusch998).
Among the bacteria, Cupriavidus necator is a well-known Gram-egative PHA producer. Hassan et al. (2002) reported that the abilityf this strain to produce PHA using concentrated organic acidsrom palm oil mill effluent (POME). However, the toxic nature ofhe organic acids substrate caused reduced yield and productivityf PHA by C. necator, which prefers fructose substrates in nature.ence a local isolate, Comamonas sp. EB172, a bacterium isolated
rom the open digester-treated-POME has been used to convertrganic acids (acetic acid:propionic acid:n-butyric acid) from POMEnto PHA (Zakaria et al. 2008, 2010a; Mumtaz et al. 2010). Coma-
onas sp. EB172 has demonstrated a higher tolerance and a fasterptake of organic acids compared with C. necator ATCC 17699Mumtaz et al. 2010). The PHA yield obtained for C. necator ATCC7699 and Comamonas sp. EB172 was 0.15 and 0.31 g/g substrate,espectively. From the previous studies by Shi and co-workers onhe metabolic flux analysis on P(3HB) synthesis (Shi et al. 1997;hi et al. 1999), Hassan and co-workers suggested recombinantNA techniques using C. necator PHB−4 transformant or recom-inant Escherichia coli for efficient microbial fermentation. It haseen demonstrated that C. necator PHB−4 transformant harbouringhaC genes derived from various sources were excellent producersor PHA (Sudesh et al. 1998; Bhubalan et al. 2008).
In order to produce recombinant strains, the genes for PHAiosynthesis should be cloned in the new host and functionalxpression of those genes should be determined. E. coli is one ofhe most widely used hosts for the production of heterologousroteins as its genetics is far better characterised than those ofny other microorganism, hence it is a suitable host for the pro-uction of PHA. PHA production using recombinant E. coli strainsas advantages over wild-type PHA producers due to an easieranipulation for higher productivity, lack of native degradationachinery and fragility of the cell membranes, which facilitates an
asy purification and recovery process (Aldor and Keasling 2003).n addition, recombinant E. coli can be utilised for homopolymernd copolymer production depending on the specificity of the PHAiosynthesis gene to the substrates used. Matsusaki et al. (2000)emonstrated that PHA-negative mutants of Pseudomonas putidaPp104 and C. necator expressing phaC1 gene from Pseudomonas sp.1-3 could produce copolymer consisting of monomers containing–12 carbon atoms. In this study, E. coli was used as a host strain toetermine the functionality of the cloned PHA biosynthesis operonrom Comamonas sp. EB172.
This paper reports the first genetic study on the PHA biosynthe-is operon of the acid-tolerant Comamonas sp. EB172 and describeshe cloning and expression of these genes in E. coli JM109 using glu-ose and/or mixed organic acids as sole carbon source. The successf cloning and expression of novel phaCABCo in E. coli will definitelyreate another option for the potential strain for the industrial PHAroduction from biomass resources in future.
aterials and methods
acterial strains, plasmids and cultivation conditions of the cells
The bacterial strains and plasmids used in this study are listedn Table 1. Comamonas sp. EB172 is a new local isolate obtained
rom POME sludge in the open digester (Zakaria et al. 2008). Coma-onas sp. EB172 was cultivated in nutrient broth (NB) at 30 ◦C
nd 200 rpm. E. coli JM109 was cultivated at 37 ◦C in Luria–Bertaniedium, containing 10 g casein peptone, 10 g sodium chloride and
5 g yeast extract in 1 L of distilled water. When necessary, the antibi-otic ampicillin (100 �g/mL) was added to the medium to maintainthe stability of the plasmids.
Amplification of phaCABCo genes from Comamonas sp. EB172
Genomic DNA was extracted from an overnight cell culture ofComamonas sp. EB172 using a GeneJETTM Genomic DNA PurificationKit (Fermentas, Glen Burnie, MD, USA). Partial sequences of phaABCogenes were isolated using degenerate primers (FA: 5′-CGG ACC GCCGTG GGN AAR TTY GG-3′; RA: 5′-CGA TGG CGC CGC CRT TNA CRTT-3′; FB: 5′-TGG CCT ACG TGA CCG GNG GNA T-3′; RB: 5′-CGC CGTTCA CGG AGA ART CNG CNC C-3′). Full-length phaABCo genes wereobtained using a DNA walking SpeedUpTM Premix Kit II (Seegene,Rockville, MD, USA). The PCR conditions for both phaACo and phaBCoamplifications were as follows: denaturation at 94 ◦C for 1 min,annealing at 66.1 ◦C for 1 min and elongation at 72 ◦C for 1 min.In addition, partial sequences of phaCCo were detected with twosets of degenerate PCR primers and the subsequent sequences wereobtained using a DNA walking SpeedUpTM Premix Kit II. However,the upstream (5′) sequence of phaCCo was identified by 5′ RACE PCR(Clontech, Mountain View, CA, USA). The first PCR using degenerateprimers (FC1: 5′-GGG CGC CGT GGT TWY GAR AAY GA-3′ and RC1:5′-AGG AGG GTT GAT GAC GCC NGC NAY RTG-3′) amplified about900 bp of good sequences. However, more than 1000 bp of phaCCosequences were obtained through the following PCR amplificationusing degenerate primers (FC1: 5′-GGG CGC CGT GGT TWY GAR AAYGA-3′ and RC2: 5′-CGA TGG CGC CGC CRT TNA CRT T-3′): denatu-ration at 94 ◦C for 1 min, annealing at 66 ◦C and 62.7 ◦C, for the firstand seconds sets of primers, respectively, for 1 min and elongationat 72 ◦C for 1 min. A DNA walking SpeedUpTM Premix Kit II and 5′
RACE PCR was completed according to the manufacturers’ proto-
tion of the vectors into E. coli JM109 prior to DNA sequencing. Thenucleotide sequences of phaCCo, phaACo and phaBCo reported herewere deposited in the GenBank online database under accessionnumbers JF773394, HQ650140 and HQ650141, respectively.
5 ical Re
C
dcfppwp(iwpsaugoccpPtpPfuwgpptccr
FP
pLoilKEwsfpPPpwbmttmpoa(
52 L.-N. Yee et al. / Microbiolog
onstruction of the plasmids and cloning of the phaCABCo genes
Specific primers for the amplification of phaCABCo wereesigned according to the sequences obtained. The oligonu-leotide primers for the amplification of the phaCCo generagment were P1: 5′-CGCCAAGAACGCAAGGATGACA-3′ (forwardrimer) and P2: 5′-GCTATGACTTGGGTGAGCGTGGT-3′ (reverserimer). The primers designed for the amplification of phaABCoere P3: 5′-GGCGGTACCCACGCTCACCCAAGTCATAGC-3′ (forwardrimer) and P4: 5′-GATGGATCCGCCGCTTTCATCTTCCATGCT-3′
reverse primer). A modified plasmid (pGEM′phbCABRe) contain-ng phbCABRe from C. necator, as reported in Bhubalan et al. (2011)
as used for gene expression in E. coli JM109. The modified plasmidGEM′phbCABRe was digested using Csp45I and PstI to remove theynthase (phbCRe) gene and was dephosphorylated using shrimplkaline phosphatase (SAP) (Promega, Madison, WI, USA). A PCRsing the respective primers for the amplification of phaCABCoenes from the genomic DNA of Comamonas sp. EB172 was carriedut to obtain the fragments for ligation. The phaCCo fragment wasonverted to blunt-ends using End-Repair of the Insert DNA (Epi-entre, Madison, WI, USA) prior to ligation to form a new plasmidGEM′-phaCCoABRe. In addition, pGEM′-phbCABRe was digested bystI and NdeI to remove the phbABRe genes from the modified vec-or, dephosphorylated with SAP and phosphorylated PCR amplifiedhaABCo was ligated to form pGEM′-phaCReABCo (Fig. 1). By using1 and P4 primers, full length of phaCABCo genes was amplifiedrom the genomic DNA of Comamonas sp. EB172. The PCR prod-ct of phaCABCo operon was phosphorylated. The pGEM′-phbCABReas digested using Csp45I and NdeI to eliminate the phbCABRe
enes and dephosphorylated with SAP. New plasmid named asGEM′-phaCABCo with entire phaCABCo operon was introduced intoGEM′ modified vector after ligation step (Fig. 1). All plasmids wereransformed using a heat shock method into E. coli JM109. Theolonies were screened on Luria–Bertani agar containing ampi-illin (100 �g/mL), and the inserts were confirmed through PCR andestriction enzyme digestion.
unctional expression of the phaCABCo genes in E. coli JM109 forHA production
E. coli JM109 transformant containing the pGEM′-phaCCoABRe,GEM′-phaCReABCo or pGEM′-phaCABCo plasmid was first grown inuria–Bertani medium to reach an optical density at the wavelengthf 600 nm of 3.0 (OD600 nm = 3.0). Subsequently, 10% (v/v) of thenoculum was transferred into 500 mL flasks with 100 mL nitrogen-imited mineral salts medium (MS) containing 5 g KH2PO4, 2 g2HPO4, 0.4 g MgSO4·7H2O and 1 mL trace element solution. These. coli JM109 transformants harbouring three different plasmidsere cultivated on 10 g/L filter-sterilised glucose as carbon sub-
trate at 37 ◦C and 200 rpm for 48 h. Fermentation to determine theunctionality of the E. coli JM109 transformants carrying 3 differentlasmids, nitrogen limiting medium was used to only confirm the(3HB) accumulation from the effect of carbon source. However, for(3HB) production, E. coli JM109 transformant harbouring pGEM′-haCABCo plasmid was grown in MS medium that supplementedith 1 g/L (NH4)2SO4 as nitrogen source for cell growth. The car-
on source, 10 and 20 g/L of filter-sterilised glucose, 5 and 10 g/L ofixed organic acids (acetic acid:propionic acid:n-butyric acid) in
he mass ratio of 3:1:1, respectively to stimulate the anaerobicallyreated POME as sole carbon source were added separately to the
edium. Ampicillin (100 �g/mL) was added when necessary for
lasmid maintenance. The grown cells were harvested after 48 hf cultivation by centrifugation (6000 × g, 10 min, 4 ◦C), lyophilisednd cells were subjected to methanolysis for gas chromatographyGC) analysis to determine the PHA content (Braunegg et al. 1978).
search 167 (2012) 550– 557
Results
Cloning of phaCABCo genes from Comamonas sp. EB172
For identification of the PHA biosynthesis genes of Comamonassp. EB172, a partial fragment of the phaCCo gene (971 bp) was ampli-fied from the genomic DNA using degenerate primers. Anotherreverse degenerate primer was used to amplify the total phaCCofragment (1161 bp), while 5′ RACE PCR was used to detect thesequences upstream of the phaCCo gene. These degenerate primerswere designed based on the conserved sequences of the phaC genesfrom the database after ClustalW alignment. The main differencebetween the 5′ RACE PCR compared to normal PCR is the geneticmaterial used. Total RNA was extracted and used as template forcomplementary DNA (cDNA) synthesis, and this synthesised cDNAwas used as a template for RACE PCR using gene specific primersthat were designed based on the known sequences of the phaCCogenes. Based on similar genes in the database, the phaA genes werefound to be clustered with phaC genes. In addition, full genomesequences from database showed some microorganisms carry thePHA biosynthesis genes in a single operon, such as Delftia acidovo-rans SPH-1, Comamonas testosteroni CNB-2 and Acidovorax sp. JS42.With this information and the sequences, we successfully amplifiedthe partial products of the phaACo and phaBCo genes using degen-erate primers which were 1035 bp and 710 bp, respectively. It wasfound that the phaACo and phaBCo genes were located downstreamof the phaCCo gene. The DNA walking method was then used toidentify the full sequences of the phaACo and phaBCo genes from thegenomic DNA of Comamonas sp. EB172. Fig. 2 shows the molecularorganisation of phaCABCo operon from Comamonas sp. EB172.
Alignment of the phaCABCo genes from Comamonas sp. EB172
The PHA biosynthesis genes of Comamonas sp. EB172 wereclustered together in one operon as phaCCo–phaACo–phaBCo. Basedon the BLAST results of the nucleotide sequences from the Gen-Bank database, the phaCCo, phaACo and phaBCo genes consisted of1694 bp, 1182 bp and 738 bp nucleotides sequences, respectively.These genes were identified as poly-(R)-hydroxyalkanoic acid syn-thase (class I), acetyl-CoA acetyltransferase and acetoacetyl-CoAreductase. The phaCCo gene encoded a protein of 564 amino acids,and the calculated molecular mass was 62.94 kDa. The 1182 bpphaACo gene encoded a protein of 394 amino acids with a calculatedmolecular mass of 40.73 kDa, while the 738 bp phaBCo gene encodeda protein of 246 amino acids with a calculated molecular mass of26.27 kDa. The phaACo and phaBCo genes exhibited homologies of27% and 28%, respectively, to those belonging to D. acidovorans SPH-1, while the phaC gene exhibited 22% homology to that of Acidovoraxebreus TPSY.
Comparison of the deduced amino acid sequences of phaCCo,phaACo and phaBCo in this study revealed significant homology tothose of closely related microorganisms. Based on the ClustalWaligned sequences, the phaCCo product from Comamonas sp. strainEB172 DSM 23953 showed 66.6%, 69.3% and 69.5% identities withthose of C. testosteroni CNB-2, Acidovorax sp. JS42 and A. ebreusTPSY, respectively. In addition, the deduced amino acid sequencesfor phaACo and phaBCo exhibited homology to those from A. ebreusTPSY, C. acidovorans and D. acidovorans SPH-1. The phaACo homol-ogy was 84.4%, 85.2% and 85.4%, respectively. In addition, the phaBCogene revealed 88.5%, 89.3% and 89.7% homology with those fromAcidovorax citrulli AAC00-1, D. acidovorans SPH-1 and Acidovoraxsp. JS42 strains, respectively. Fig. 3 shows the ClustalW alignments
of several similar phaCCo sequences of the PHA biosynthesis genesfrom the GenBank database with those from Comamonas sp. EB172.The clustered phaCCo sequences showed that a fragment (about123 bp) located after the lipase box-like sequence (G-F-C-V-G-G)
L.-N. Yee et al. / Microbiological Research 167 (2012) 550– 557 553
Fig. 1. Construction of the pGEM′-phbCABRe plasmid for expression of the phaCCo , phaABC
from C. necator.
1694 bp 1182 bp 738 bp
PCo phaC pha A pha B
FEa
123 bp 193 bp
ig. 2. Molecular organisation of phaCABCo biosynthesis genes from Comamonas sp.B172 is presented. PHA synthase (phaCCo), acetyl-CoA acetyltransferase (phaACo),cetoacetyl-CoA reductase (phaBCo) genes and promoter (PCo).
o and phaCABCo genes in Escherichia coli JM109. Promoter (PRe) and terminator (TRe)
which was not homologous to the phaC from Comamonas or Delftia.The region between the codon GAG, coding for glutamic acid (E),and GAT, coding for aspartic acid (D), contributed to the remark-able differences among the isolated phaCCo from C. acidovorans, C.testosteroni CNB-2 and D. acidovorans SPH-1. However, the align-ment for phaACo and phaBCo showed a high degree of homology torelated sequences from the GenBank database. From the alignment
of the several different phaA genes from the database, the detectionof 2 highly conserved cysteine residues at positions 88 and 378 wasrevealed. A histidine residue at position 348 was also obtained fromthe highly identical region which similar as previously reported
554 L.-N. Yee et al. / Microbiological Research 167 (2012) 550– 557
1 ---------- ----MNSTPDWGAAGEQYTDFFSRQWQNAFQSFAALGQAEPAKT---------------- 40 C. sp . EB172 1 ---------- -----------------MSTNFDASWADSARQFQQMVGEGWSQMLQVLQPKEMG------ 37 A. eb reus TPSY 1 MRCSFLRHRL LGNNDTWGPRLGRDGGCMSTNFDASWADSARQFQQMVGEGWSQMLQVLQPKEMG------ 64 A. sp . JS42 1 ---------- -----------------------------------MSG---------------------- 3 Alc. l atus 1 ---------- ----MNFDPLAGLSGQSVQQFWNEQWSRTLQTLQQMGQPGLPGIQGMPGMPDMAQAWK-- 54 C. ac idovorans 1 ---------- ----MNFDPLAGLSGQSVQQFWNEQWSRTLQTLQQLGQPGLPGTQGMPGMPDMAQAWK-- 54 D. ac idovorans SPH-1 1 ---------- ----MNFDPSWGQAGQSIQQFWQEQWSKSLQTLQQMQSGAPMGLSAMPGLGAAPNPWAGI 56 C. te stosteroni CNB-2 40 -----------------AVQLDVHRLQA LQQEYVQEATRLWSQG-------LQAPAQLQ -DRRFKSA PWL 85 C. s p. EB172 37 -----AALSMPAAG---PVGFAPDKLAA VQQRYLQEVQALWSHAL------HHDDAPLS NDRRFAGE NWA 93 A. e breus TPSY 64 -----AALSMPAAG---PVSFAPDKLAA VQQRYLQEVQALWSHAL------HHDEAPLS NDRRFAGE NWA 120 A. sp. JS42 3 ---------- --------LNLPMQAMTK LQGEYLNEATALWNQTLGRLQPDGSAQPAKL GDRRFSAE DWA 55 Alc. latus 54 -----AAVPEPGALPENALSLDPEKLLE LQRQYLDGAKAMAEQ--------GGAQALLA KDKRFNTE SWA 111 C. acidovorans 54 -----AAVPEPGALPENALSLDPEKLLE LQRQYLDGAKAMAEQ--------GGAQALLA KDKRFNTE SWA 111 D. acidovorans SPH-1 57 MDAMQSAMPQMATGSAQAVQFDAAKLQG LQQEYLQSVQSLADG--------KQIQALLS RDKRFAKP EWS 118 C. testosteroni CNB-2 86 SNPVAAMTA SAYLLNAK ALMGMADAVQ ADEKTR QRIRFAV EQWVAAMSPS NYLALNPEAQQKALETKGE S 155 C. sp. EB172 94 HNPLSAFSV AAYQLQAH ALMGLADAVQ ADDKTR ARIRFSV EQWLAAMAPS NFLVFNADAQQKALETHGE S 163 A. ebreus TPSY 121 HNPLSAFSVA AYQLQAH ALMGLVDAVQ ADDKTR ARIRFSV EQWLAAMAPS NFLVFNADAQQKALETHGE S 190 A. sp. JS42 56 KNPAAAYLA QVYLLNAR TLMQMAESIE GDAKAK ARVRFAV QQWIDAAAPS NFLALNPEAQRKALETKGE S 125 Alc. latus 112 GNPLTAATAA TYLLNSR MLMGLADAVQ ADDKTR NRVRFAI EQWLAAMAPS NFLALNAEAQKKAIETQGE S 181 C. acidovorans 112 GNPLTAATAA TYLLNSR MLMGLADAVQ ADDKTR NRVRFAI EQWLAAMAPS NFLALNAEAQKKAIETQGE S 181 D. acidovorans SPH-1 119 ANPVASMAAA NYLLGSR MLTGMAEAVQ GDEKTR NRVRFAV EQWVAAMAPS NFLALNADALNKVVETKGE S 188 C. testosteroni CNB-2 156 LAKGIQNML HDLQQG HLSMTDE SAFEVGKNVATTEGAV VYENELFQ LIEYKPL TGKVYK RPLLLVPPCI N 225 C. sp. EB172 164 IAKGVANLL HDLGQG HISMTDE SRFEVGRNVATTEGAV VFENELFQ LLEYKPL TPKVYE RPLLMVPPCI N 233 A. ebreus TPSY 191 IAKGVANLL HDLGQG HISMTDE SRFEVGRNVATTEGAV VFENELFQ LLEYKPL TAKVYE RPLLMVPPCI N 260 A. sp. JS42 126 ISQGLQQLW HDIQQG HVSQTDE SVFEVGKNVATTEGAV VYENDLFQ LIEYKPL TPKVHE KPMLFVPPCI N 195 Alc. latus 182 LAQGVANLL ADMRQG HVSMTDE SLFTVGKNVATTEGAV VFENELFQ LIEYKPL TDKVHE RPFLMVPPCI N 251 C. acidovorans 182 LAQGVANLL ADMRQG HVSMTDE SLFTVGKNVATTEGAV VFENELFQ LIEYKPL TDKVHE RPFLMVPPCI N 251 D. acidovorans SPH-1 189 LAQGIANLL ADMRQG HVSMTDE SLFTVGKNVATTEGAV VFENELFQ LIEYKPL TAKVFE KPFLMVPPCI N 258 C. testosteroni CNB-2 226 KYYILDLQ PENSLIRYAVSQGH RTFVVSWRNP DESLDHL GWDDYIED AVIRAIDVTQ QIAGADQ INTLG F 295 C. sp. EB172 234 KYYILDLQ PDNSLIRHAVAQGH RTFVVSWRNP DASLAHK TWDDYIDE AVLTAIATVQ QIAGAKQ INALG F 303 A. ebreus TPSY 261 KYYILDLQ PDNSLIRHAVAQGH RTFVVSWRNP DASLAHK TWDDYIDE AVLTAIATVQ QIAGAKQ INALG F 330 A. sp. JS42 196 KYYILDLQ PDNSLIRYTVAQGH RVFVVSWRNP DASVAGK TWDDYVEQ GVIRAIRVMQ QITGHEK VNALG F 265 Alc. latus 252 KFYILDLQ PDNSLIRYAVSQGH RTFVMSWRNP DESLARK TWDNYIED GVLTGIRVAR EIAGAEQ INVLG F 321 C. acidovorans 252 KFYILDLQ PDNSLIRYAVSQGH RTFVMSWRNP DESLARK TWDNYIED GVLTGIRVAR EIAGAEQ INVLG F 321 D. acidovorans SPH-1 259 KFYILDLQ PDNSVIRYAVSQGH RTFVVSWRNP DESQAHK SWDDYIED GVLKAVSTVQ DICEAPQ INVLG F 328 C. testosteroni CNB-2 296 CVGG TMLATALAV LAARGE----------------------------------------- DPVASATLLT 324 C. sp. EB172 304 CVGG TMLATALAV LAARGE----------------------------------------- EPVASCTFLT 332 A. ebreus TPSY 331 CVGG TMLATALAV LAARGE----------------------------------------- EPVASCTFLT 359 A. sp. JS42 266 CVGG TILSTALAV LAARGE----------------------------------------- QPAASLTLLT 294 Alc. latus 322 CVGG TMLSTALAV LQARHDREHGAVAAP-AAKAPAAKRAAGSRSAARTSTARATAPAGVP FPVASVTLLT 390 C. acidovorans 322 CVGG TMLSTALAV LQARHDREHGAVAAPAAAKAPAAKRAAGSRSAARTSTARATAPAGVP FPVASVTLLT 391 D. acidovorans SPH-1 329 CVGG TMLSTAMAV LAARHAREHG---QPQAAKP--RSRAAAAKTGSKTAAAKVEAHA--- FPVASMTLLT 390 C. testosteroni CNB-2 325 TFVDFSSTGVLDVFID EAFVQMREMQ MGK----- GGLLKGQE LAATFSFLRPNDLVWNYVVGNYLKG ETP 389 C. sp. EB172 333 TLIDFADTGILDVFID EGFVRMRELQ MGQ----- GGLMKGQD LASTFSFLRPNDLVWNYVVGNYLKG ETP 397 A. ebreus TPSY 360 TLIDFADTGILDVFID EGFVRMRELQ MGQ----- GGLMKGQD LASTFSFLRPNDLVWNYVVGNYLKG ETP 424 A. sp. JS42 295 TLLDFSNTGVLDLFID EAGVRLREMT IGEKAPNG PGLLNGKE LATTFSFLRPNDLVWNYVVGNYLKG EAP 364 Alc. latus 391 TFIDFSDTGILDVFID ESVVRFREMQ MGE----- GGLMKGQD LASTFSFLRPNDLVWNYVVGNYLKG ETP 455 C. acidovorans 392 TFIDFSDTGILDVFID ESVVRFREMQ MGE----- GGLMKGQD LASTFSFLRPNDLVWNYVVGNYLKG ETP 456 D. acidovorans SPH-1 391 TLIDFRDTGILDIFID ENMVRLREMQ MGK----- GGLMKGQD MASTFSFLRPNDLVWNYVVGNYLKG ETP 455 C. testosteroni CNB-2 390 PPFDLLYWNS DATNLPGPWY AWYLR NTYLE NNLIRPGKLTVCG EQIDMGRVT LPLYIYGSREDHIV PIEA 459 C. sp. EB172 398 PPFDLLYWNS DSTNLPGPYY AWYLR NFYLE NNLIRPGRLTVCG ESLDLTQVQ LPVYIYGSREDHIV PIGA 467 A. ebreus TPSY 425 PPFDLLYWNS DSTNLPGPYY AWYLR NFYLE NNLIRPGRLTVCG ESLDLTQVQ LPVYIYGSREDHIV PIGA 494 A. sp. JS42 365 PPFDLLYWNS DSTNMAGPMF CWYLR NTYLE NKLRVPGALTICG EKVDLSRIE APVYFYGSREDHIV PWES 434 Alc. latus 456 PPFDLLYWNS DSTNLPGPYY AWYLR NLYLE NRLAQPGALTVCG ERIDMHQLR LPAYIYGSREDHIV PVGG 525 C. acidovorans 457 PPFDLLYWNS DSTNLPGPYY AWYLR NLYLE NKLAQPGALTVCG ERIDMHQLR LPAYIYGSREDHIV PVGG 526 D. acidovorans SPH-1 456 PPFDLLYWNS DSTNLPGPFY AWYLR NLYLE NNLIKPGVLTVCG EKIDMSQLK MPVYIYGSREDHIV PVAA 525 C. testosteroni CNB-2 460 AYASTQVF PGDKRFVMGASGHIAGVINP PAKHKR SHWIRSDNALPASHG EWLQEA TEHPGSW WTDWSQWL 529 C. sp. EB172 468 AYASTQVL PGDKRFVMGASGHIAGVINP PAKKKR SHWLREDGQLPATLD AWLAGA TELPGSW WDDWCAWL 537 A. ebreus TPSY 495 AYASTQVL PGDKRFVMGASGHIAGVINP PAKKKR SHWLREDGQLPATLD AWLAGA TELPGSW WDDWCAWL 564 A. sp. JS42 435 AYAGTQML SGPKRYVLGASGHIAGVINP PQKKKR SYWTN--EQLDGDFN QWLEGS TEHPGSW WTDWSDWL 502 Alc. latus 526 SYASTQVL GGDKRFVMGASGHIAGVINP PAKKKR SYWLREDGQLPATLK EWQAGA DEYPGSW WADWSPWL 595 C. acidovorans 527 SYASTQVL GGDKRFVMGASGHIAGVINP PAKKKR SYWLREDGQLPPTLK EWQAGA DEYPGSW WADWSPWL 596 D. acidovorans SPH-1 526 AYASSQVL GGERRFVMGASGHIAGVINP PAKNKR SHWLREDGEFPEAFN DWLAEA TEYPGSW WTDWSAWL 595 C. testosteroni CNB-2 530 QSHAGTQ IAAP KRYGKGRSYEA LEPAPGRY VREKA 564 C. sp. EB1 72 538 HGHAGRQ IAAP KAYGKAPKFKA IEPAPGRY VQQKA 572 A. ebreus TPSY 565 HGHAGRQ IAAP KAYGKAPKFKA IEPAPGRY VQQKA 599 A. sp. JS4 2 503 KQHAGKE IAAP KTPGN-KTHKP IEPAPGRY VKQKA 536 Alc. latu s 596 AEHGGKL VAAP KQYGKGREYTA IEPAPGRY VLVKA 630 C. acidovo rans 597 AEHGGKL VAAP KQYGKGREYTA IEPAPGRY VLVKA 631 D. acidovo rans SPH-1 596 GSHAGKQ LAAP KSYGRARAYEA IEDAPGRY VLSKA 630 C. testost eroni CNB-2
Lip ase box-like
Fig. 3. Alignment of the deduced amino acid sequence of the PHA synthase from Comamonas sp. EB172 with those from the following strains: Acidovorax ebreus TPSY(CP001392), Acidovorax sp. JS42 (CP000539), Alcaligenes latus(AF078795), Comamonas acidovorans (AB009273), Delftia acidovorans SPH-1 (CP000884) and Comamonas testos-teroni CNB-2 (CP001220). Amino acids that are identical between all the PHA synthases are identified by black boxes whereas conserved and similar amino acids are highlightedin light grey.
ical Re
(s
H
pcftscdPoTicphtlJ1aEtc4
giPpmhC(tamw
vt1digrmflCsTg
D
wsab
L.-N. Yee et al. / Microbiolog
Lindenkamp et al. 2010). These 3 amino acids involve in the activeite cavity formation in the PHA biosynthesis pathway.
eterologous expression of phaCABCo genes in E. coli JM109
The phaCCo, phaACo and phaBCo genes were expressed usingGEM′, which is a modified version of the commercially availableloning vector, pGEM-T (Promega), in E. coli JM109 to determine theunctionality of these genes. P(3HB) production by the E. coli JM109ransformants carrying the different plasmids with PHA biosynthe-is genes was evaluated after 48 h cultivation in minimal mediumontaining glucose (1%, w/v) and without any nitrogen source toetermine if the genes were able to confer the ability to synthesiseHA (Table 2). Since glucose was used as the sole carbon source,nly P(3HB) homopolymer was synthesised by the transformants.he P(3HB) accumulation by E. coli JM109 transformants harbour-ng different plasmids varied from 17 wt.% to 46 wt.% PHA in theellular dry mass. E. coli JM109 transformants harbouring pGEM′-haCCoABRe showed better expression of genes by synthesisingigher P(3HB) compared to pGEM′-phaCReABCo. The E. coli JM109ransformant harbouring pGEM′-phaCCoABRe was able to accumu-ate PHA up to 33 wt.% of cellular dry mass. However, the E. coliM109 transformant harbouring pGEM′-phaCReABCo only produced7 wt.% of P(3HB). From Table 2, we could see that the phaCCond phaABCo genes were heterologously expressed. In addition, the. coli JM109 transformants harbouring pGEM′, which contains onlyhe single operon phaCABCo, produced P(3HB). About 1.06 g/L ofell dry weight (CDW) was obtained and the P(3HB) content was1 wt.% of CDW.
Table 3 shows a comparison of the P(3HB) accumulation usinglucose as sole carbon source in E. coli JM109 transformant harbour-ng different PHA biosynthesis gene from various microorganisms.(3HB) accumulation by E. coli JM109 transformants harbouringGEM′-phaCABCo are comparable to those E. coli JM109 transfor-ant with different phaCAB genes in shake flask fermentation. The
eterologous expression of phaCABCo operon was controlled by the. necator promoter and isopropyl-�-d-1-thiogalactopyranosideIPTG) was not added throughout the experiments. In addition,his E. coli JM109 transformants harbouring pGEM′-phaCABCo canccumulate P(3HB) from glucose compared to the wild type Coma-onas sp. EB172 where minimal growth and PHA accumulationas observed (Zakaria et al. 2010a,b).
In order to investigate the ability of the phaCABCo genes to utiliseolatile organic acids for P(3HB) accumulation, the E. coli JM109ransformants harbouring pGEM′-phaCABCo were grown with 5 or0 g/L of mixed organic acids. The amount of accumulated P(3HB)epends on the ratio of the mixed organic acids initially present
n the MS medium (Fig. 4). The maximum P(3HB) contents in cellsrown with 5 or 10 g/L mixed organic acids were 6.2% and 43.6%,espectively. E. coli JM109 transformants were capable of accu-ulating P(3HB) using glucose and also short chain fatty acids in
ermentation even though the cell growth and accumulation wasower using mixed acids compared with the parent strain/wild typeomamonas sp. EB172. The content of P(3HB) accumulation wasimilar to production of E. coli JM109 transformants using glucose.he content of P(3HB) accumulation was similar to that strain usinglucose.
iscussion
In this study, the genetic modification of E. coli JM109 together
ith the isolated phaCABCo demonstrated the broad substrate
pecificity by utilising different carbon sources (sugar based or fattycids) to accumulate the polymer. The results showed the capa-ility of the original non-PHA producer E. coli to accumulate high
search 167 (2012) 550– 557 555
amounts of polymer which was maintained without degradationduring the fermentation process. Thus, the production cost can bereduced not only based on raw materials but also the fermentationusing a naturally non-PHA degrader strain. In addition, complexmedium or advanced fermentation processes are not required forcultivation of recombinant E. coli, making it possible to develop PHAas biodegradable plastics widely used in the plastic industry. TheE. coli transformant will become fragile with high accumulationof intracellular polymer, which is advantageous for the down-stream extraction and recovery process for pure polymer. E. colitransformant is useful in PHA production for cost reduction anddownstream processing recovery and purification system, which isan ideal way to enable PHA to economically compete with conven-tional petrochemical plastics.
The key enzyme for PHA biosynthesis is PHA synthase, whichcatalyses the polymerisation of 3-hydroxyacyl-CoA substrates intothe PHA polymer (Rehm and Steinbüchel 1999). PHA synthasesbelonging to class I are active in the polymerisation of short-chain hydroxyacyl-CoA, comprising 3–5 carbon atoms. The putativemolecular weight of PhaCCo of 62 kDa represents PHA synthasesbelonging to class I with only one subunit (PhaC) (Rehm andSteinbüchel 1999; Qi and Rehm 2001). In this study, both phaACoand phaBCo genes of Comamonas sp. EB172 showed similarity tothose belonging to D. acidovorans SPH-1.
Based on the results obtained in this study, the phaCABCogenes in Comamonas sp. EB172 were present in an operon thatcan be cloned and functionally expressed in an E. coli JM109 forP(3HB) production. It is important to highlight that the presenceof phaCABCo of Comamonas sp. EB172 in a single operon was alsoreported in related microorganisms such as D. acidovorans SPH-1, C. testosteroni CNB-2 and Acidovorax sp. JS42 based on the fullgenome sequence information of these organisms. However, thePHA biosynthesis genes in single operon of C. acidovorans DS-17consisted of only phaC and phaA (Sudesh et al. 1998). In addition, thenucleotide sequences of the single operon of phaCABCo from Coma-monas sp. EB172 were highly similar to that of A. latus with about63% similarity (64% for phaC, 81% for phaA and 75% for phaB). Inthis study, we have successfully cloned and expressed the PHA syn-thase gene (phaCCo), the acetyl-CoA acetyltransferase gene (phaACo)and the acetoacetyl-CoA reductase gene (phaBCo) of Comamonassp. EB172, capable of producing P(3HB) from glucose and mixedorganic acids.
The phaCCo of Comamonas sp. EB172 was heterologously co-expressed with phbABRe of C. necator or phaABCo of Comamonassp. EB172. In addition, the latter genes were co-expressed withphbCRe from C. necator. Different E. coli JM109 transformants wereconstructed to show the ability or level of expression of the iso-lated genes for P(3HB) accumulation. E. coli JM109 transformantsharbouring pGEM′-phaCCoABRe and pGEM′-phaCReABCo producedP(3HB) under the control of the promoter for C. necator, suggest-ing that the isolated genes are able to be expressed by couplingwith different PHA biosynthesis genes in the heterologous host.Heterologous expression of these genes in E. coli JM109 resulted inthe accumulation of P(3HB), which indicated that the phaCCo andphaABCo genes are functionally active in the P(3HB) biosynthesispathway. Although all the E. coli JM109 transformants showed thelower accumulation of P(3HB) compared to the E. coli JM109 trans-formant harbouring pGEM′-phbCABRe of C. necator (Table 2), butthe P(3HB) content produced by the transformant with only thephaCABCo operon was comparable to that of the recombinant withphbCABRe from C. necator under nitrogen limiting condition with 1%(w/v) glucose as carbon source.
The successful expression of phaCABCo in E. coli JM109 trans-formant under the control of promoter of C. necator with glucoseand mixed organic acids will contribute towards developing a lab-oratory or industrial scale production of P(3HB) and even promote
556 L.-N. Yee et al. / Microbiological Research 167 (2012) 550– 557
Table 2P(3HB) accumulation in E. coli JM109 transformant strains.a
Data shown are the average of duplicate analyses and standard deviations are presented.a Cells were grown using a two-stage cultivation process in minimal medium containing 1% (w/v) glucose as a carbon source and without any nitrogen source. E. coli JM109
transformant harbouring pGEM′-phbCABRe was grown as positive control in the experiments.b The P(3HB) content in lyophilised cells was analysed using gas chromatography.
Table 3Comparison of P(3HB) accumulation using glucose in E. coli with different PHA biosynthesis genes.
Microorganism Insert Substrate, culture mode P(3HB) content (wt.% of CDW)References
ig. 4. Time profile of the (A) cell dry weight of E. coli JM109 transformants harbourcids. Data shown are the average of duplicate analysis.
opolymer production. When glucoses were used for PHA pro-uction, E. coli JM109 transformant harbouring PHA biosynthesisenes of Comamonas sp. EB172 (phaCABCo) performed better com-ared to wild type Comamonas sp. EB172. However, only P(3HB)ith trace amount of 3HV could be produced from the mixed
rganic acids. This may be due to the simultaneous presence ofoth acetic and propionic acids as potential inducers of propionateetabolic pathway for the 3-HV fraction in the accumulated poly-ers (Slater et al. 1992). Choi and Lee (1999) demonstrated the high
oncentration of poly(3-hydroxybutyrate-co-3-hydroxyvalerate)P(3HB-co-3HV)] could be synthesised by fatty acid induction oreeding strategy. At higher concentration (10 g/L) of organic acids, arace amount of 3-HV (less than 1%) was produced once the fermen-ation reached 48 h (data not shown). Thus, our further research willocus on copolymer accumulation and improvement in cell den-ity when using organic acids as carbon substrates. The different
ermentation strategies such as fermentation medium, conditionsnd feeding strategy with glucose or organic acids should be inves-igated to get higher total biomass and polymer accumulation inhorter cultivation time using cost-effective carbon sources.
EM -phaCABCo and (B) P(3HB) accumulation from 5 g/L and 10 g/L of mixed organic
In this work, E. coli JM109 transformants were designed todemonstrate that phaCABCo was functional. Such heterologousexpression of phaCABCo genes represented a significant steptowards better PHA production and cell titre during growth inbioreactors using sugars or unfavourable fatty acids from wastestreams as carbon feedstocks. The utilisation of waste materialsfor PHA biosynthesis is a good strategy for cost-efficient biopoly-mer production which can also help the industry to overcome theirdisposal problems. However, the application of the biopolymersproduced need to be further studied. The performance of PHA pro-duction using genetically modified Comamonas sp. EB172 will beaddressed in our further research.
Acknowledgements
The authors would like to acknowledge the financial supportprovided by the Ministry of Science, Technology and Innovation,Malaysia and Universiti Putra Malaysia. We are grateful for thesuggestions provided by Kesaven Bhubalan from Universiti Sains
ical Re
MC
A
t
R
A
A
B
B
B
B
C
C
C
H
K
L
L
L
L.-N. Yee et al. / Microbiolog
alaysia and Tabassum Mumtaz from Bangladesh Atomic Energyommission, Bangladesh.
ppendix A. Supplementary data
Supplementary data associated with this article can be found, inhe online version, at doi:10.1016/j.micres.2011.12.006.
eferences
ldor IS, Keasling JD. Process design for microbial plastic factories: metabolicengineering of polyhydroxyalkanoates. Curr Opin Biotechnol 2003;14:475–83.
hubalan K, Chuah JA, Shozui F, Brigham CJ, Taguchi S, Sinskey AJ, Rha CK, SudeshK. Characterization of the highly active polyhydroxyalkanoate synthase of Chro-mobacterium sp. strain USM2. Appl Environ Microbiol 2011;77:2926–33.
hubalan K, Lee WH, Loo CH, Yamamoto T, Tsuge T, Doi Y,Sudesh K. Controlled biosynthesis and characterization ofpoly(3-hydroxybutyrate-co-3-hydroxyvalerate-co-3-hydroxyhexanoate)from mixtures of palm kernel oil and 3HV precursors. Polym Degrad Stab2008;93:17–23.
raunegg G, Sonnleitner B, Lafferty RM. A rapid gas chromatographic method for thedetermination of poly-�-hydroxybutyric acid in microbial biomass. Eur J ApplMicrobiol 1978;6:29–37.
hien CC, Hong CC, Soo PC, Wei YH, Chen SY, Cheng ML, Sun YM. Func-tional expression of phaCAB genes from Cupriavidus taiwanensis strain 184 inEscherichia coli for polyhydroxybutyrate production. Appl Biochem Biotechnol2010;162:2355–64.
hoi JI, Lee SY. High-level production of poly(3-hydroxybutyrate-co-3-hydroxyvalerate) by fed-batch culture of recombinant Escherichia coli.Appl Environ Microbiol 1999;65:4363–8.
hoi JI, Lee SY, Han K. Cloning of the Alcaligenes latus polyhydroxyalkanoatebiosynthesis genes and use of these genes for enhanced production ofpoly(3-hydroxybutyrate) in Escherichia coli. Appl Environ Microbiol 1998;64:4897–903.
assan MA, Nawata O, Shirai Y, Rahman NAA, Yee PL, Ariff AB, Karim MIA. A pro-posal for zero emission from palm oil industry incorporating the production ofpolyhydroxyalkanoates from palm oil mill effluent. J Chem Eng Jpn 2002;35:9–14.
hanna S, Srivastava AK. Recent advances in microbial polyhydroxyalkanoates. Pro-cess Biochem 2005;40:607–19.
i R, Zhang H, Qi Q. The production of polyhydroxyalkanoates in recombinantEscherichia coli. Bioresour Technol 2007;98:2313–20.
i ZJ, Cai L, Wu Q, Chen GQ. Overexpression of NAD kinase in recom-binant Escherichia coli harbouring the phbCAB operon improves poly(3-
indenkamp N, Peplinski K, Volodina E, Ehrenreich A, Steinbüchel A. Impact ofmultiple �-ketothiolase deletion mutations in Ralstonia eutropha H16 on thecomposition of 3-mercaptopropionic acid-containing copolymers. Appl EnvironMicrobiol 2010;76:5373–82.
search 167 (2012) 550– 557 557
Matsumoto K, Yamada M, Leong CR, Jo SJ, Kuzuyama T, Taguchi S. A new pathwayfor poly(3-hydroxybutyrate) production in Escherichia coli and Corynebacteriumglutamicum by functional expression of a new acetoacetyl-coenzyme A synthase.Biosci Biotechnol Biochem 2011;75:364–6.
Matsusaki H, Abe H, Taguchi K, Fukui T, Doi Y. Biosynthesis of poly(3-hydroxybutyrate-co-3-hydroxyalkanoates) by recombinant bacteria expressingthe PHA synthase gene phaC1 from Pseudomonas sp61-3. Appl Microbiol Biotech-nol 2000;53:401–9.
Mumtaz T, Yahaya NA, Abd-Aziz S, Nor Aini AR, Yee PL, Shirai Y, Hassan MA. Turningwaste to wealth-biodegradable plastics polyhydroxyalkanoates from palm oilmill effluent – a Malaysian perspective. J Cleaner Prod 2010;18:1393–402.
Naik S, Gopal SKV, Somal P. Bioproduction of polyhydroxyalkanoates from bacteria:a metabolic approach. World J Microbiol Biotechnol 2008;24:2307–14.
Qi Q, Rehm BHA. Polyhydroxybutyrate biosynthesis in Caulobacter crescentus:molecular characterization of the polyhydroxybutyrate synthase. Microbiology2001;147:3353–8.
Rehm BHA. Polyester synthases: natural catalysts for plastics. Biochem J2003;376:15–33.
Rehm BHA, Steinbüchel A. Biochemical genetic analysis of PHA synthases and otherproteins required for PHA synthesis. Int J Biol Macromol 1999;25:3–19.
Shi H, Nakawa J, Shimizu K. Effect of modifying metabolic network on poly-3-hydroxybutyrate biosynthesis in recombinant Escherichia coli. J Biosci Bioeng1999;87:666–77.
Shi H, Shiraishi M, Shimizu K. Metabolic flux analysis for biosynthesis of poly(�-hydroxybutyric acid) in Alcaligenes eutrophus from various carbon sources. JFerment Bioeng 1997;84:579–87.
Slater S, Gallaher T, Dennis D. Production of poly(3-hydroxybutyrate-co-3-hydroxyvalerate) in a recombinant Escherichia coli strain. Appl Environ Microbiol1992;58:1089–94.
Steinbüchel A, Füchtenbusch B. Bacterial and other biological systems for polyesterproduction. Trends Biotechnol 1998;16:419–27.
Steinbüchel A, Lütke-Eversloh T. Metabolic engineering and pathway constructionfor biotechnological production of relevant polyhydroxyalkanoates in microor-ganisms. Biochem Eng J 2003;16:81–96.
Sudesh K, Fukui T, Doi Y. Genetic analysis of Comamonas acidovorans poly-hydroxyalkanoate synthase and factors affecting the incorporation of4-hydroxybutyrate monomer. Appl Environ Microbiol 1998;64:3437–43.
Sudesh K, Iwata T. Sustainability of biobased and biodegradable plastics. Clean2008;36:433–42.
Sudesh K, Loo CY, Goh LK, Iwata T, Maeda M. The oil-absorbing property of polyhy-droxyalkanoate films and its practical application: a refreshing new outlook foran old degrading material. Macromol Biosci 2007;7:1199–205.
Umeda F, Kitano Y, Murakami Y, Yagi K, Miura Y, Mizoguchi T, Cloning. sequenceanalysis of the poly(3-hydroxyalkanoic acid)-synthesis genes of Pseudomonasacidophila. Appl Biochem Biotechnol 1998;70–72:341–52.
Verlinden RAJ, Hill DJ, Kenward MA, Williams CD, Radecka I. Bacterial synthesis ofbiodegradable polyhydroxyalkanoates. J Appl Microbiol 2007;102:1437–49.
