Appl Microbiol Biotechnol (2000) 53: 690694

Springer-Verlag 2000

ORIGINAL PAPER

D. K. Y. Solaiman R. D. Ashby T. A. Foglia

Rapid and specic identication of medium-chain-length polyhydroxyalkanoate synthase gene by polymerase chain reaction

Received: 16 August 1999 / Received revision: 23 December 1999 / Accepted: 4 January 2000

Abstract A polymerase chain reaction (PCR) protocol was developed for the specic detection of genes coding for type II polyhydroxyalkanoate (PHA) synthases. The primer-pair, I-179L and I-179R, was based on the highly conserved sequences found in the coding regions of Pseudomonas phaC1 and phaC2 genes. Puried genomic DNA or lysate of colony suspension can serve equally well as the target sample for the PCR, thus aording a simple and rapid screening of phaC1/C2-containing microorganisms. Positive samples yield a specic 540-bp PCR product representing partial coding sequences of the phaC1/C2 genes. Using the PCR method, P. corrugata 388 was identied for the rst time as a mediumchain-length (mcl)-PHA producer. Electron microscopic study and PHA isolation conrmed the production of mcl-PHA in P. corrugata 388. The mcl-PHA of this organism has a higher molecular weight than that of similar polymers produced by other pseudomonads.

Introduction
Polyhydroxyalkanoates (PHAs) are biodegradable polymers synthesized by many microorganisms (Steinbuchel 1991). With few exceptions, these polymers are usually accumulated as inclusion bodies when the organism is grown under conditions in which the carbon substrate is in excess but another nutrient component, such as nitrogen, phosphorus, sulfur or oxygen, is
Mention of brand or rm names does not constitute an endorsement by the U.S. Department of Agriculture over others of a similar nature not mentioned. D. K. Y. Solaiman (&) R. D. Ashby T. A. Foglia U.S. Department of Agriculture, Agricultural Research Service, Eastern Regional Research Center, 600 East Mermaid Lane, Wyndmoor, PA 19038, USA e-mail: dsolaiman@arserrc.gov Tel.: +1-215-2336476 Fax: +1-215-2336559

limited (Anderson and Dawes 1990). PHAs are generally grouped into two classes depending on the carbon chain length of the b-hydroxy ester monomers. Shortchain-length (scl-)PHAs contain monomer repeat-units of 35 carbon atoms, whereas medium-chain-length (mcl-)PHAs are composed of monomer repeat-units of 614 carbon atoms (Lee 1996). The genetic organization of PHA biosynthesis genes varies among PHA-producing organisms. Three major classes of PHA biosynthesis loci have been described (Poirier et al. 1995; Steinbuchel and Fuchtenbusch 1998). In the rst class (type I system), as typied by the pha locus of Ralstonia eutropha (formerly Alcaligenes eutrophus), the gene encoding the PHA synthase (phbC) is adjacent to phbA and phbB. These respectively code for b-ketothiolase and acetoacetyl-CoA reductase, two enzymes closely linked to the biosynthesis of scl-PHA. The second class (type II) of PHA genetic system consists of two synthase genes (phaC1 and phaC2) separated by the gene coding for the depolymerization of PHA (phaZ). The type II system is commonly found in mclPHA-producing pseudomonads. The type III PHA biosynthesis gene cluster is found in Chromatium vinosum, Synechocystis spp, and Thiocystis violacea. In these organisms, the synthase enzyme is composed of two polypeptide subunits encoded by phbE and phbC genes. The phbA and phbB genes are also located in this locus, but are usually transcribed divergently from the phbE and phbC genes. Recently, McCool and Cannon (1999) characterized the pha locus of Bacillus megaterium and suggested that it represented type IV PHA biosynthesis genetic organization. Methods for identifying PHA-producing organisms abound (Spiekermann et al. 1999; Takagi and Yamane 1997). The majority of these methods, however, employ lipophilic dyes to stain the polymers or cause them to uoresce. Although highly sensitive, these reagents also react with other lipid inclusion bodies and thus are not specic for PHA. Furthermore, the production of PHA is often dependent on specic growth conditions (Anderson and Dawes 1990). If such conditions are not

691

met and the polymer is not produced, then the dye-based screening would fail to identify the microorganisms as having PHA-producing capability. Timm et al. (1994) described a Southern-blot hybridization method for identifying PHA synthase genes. Their procedure allowed for broad-spectrum detection of both the phbCand the phaC-type PHA synthase genes. In this communication, we report a rapid and sensitive polymerase chain reaction (PCR) procedure for the specic detection of phaC-type genes. The procedure was tested in a screening experiment, resulting in the rst description of Pseudomonas corrugata as a mcl-PHA producing organism.

transformants with the Wizard Minipreps system (Promega). Sequencing reactions with double-stranded DNA template were carried out using an ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction kit (PE Applied Biosystems). Sequence data were acquired and analyzed on an ABI PRISM 310 Genetic Analyzer (PE Applied Biosystems). Bioinformatic analysis of DNA sequences was performed with BLAST (Altschul et al. 1997), BLAST2 (Tatiana and Madden 1999) and Omiga 1.1.3 (Oxford Molecular Group, Beaverton, Ore.) programs. PHA isolation and characterization Pseudomonas cultures were grown in 1-l Erlenmeyer asks containing 500 ml of E* medium (Brandl et al. 1988) supplemented with glucose (0.5% w/v) and/or oleic acid (0.5% v/v). Cells were grown at 30 C with 200250 rpm orbital-shaking for 72 h. When needed, an aliquot was removed and processed for transmission electron microscopic imaging as previously described (Solaiman et al. 1999). PHAs were extracted from the cells as previously described (Solaiman et al. 1999). The monomer repeat-unit composition and molecular mass of the polymer were determined by gas chromatography/mass spectroscopic analysis and gel permeation chromatography, respectively (Cromwick et al. 1996).

Materials and methods

Bacteria and growth conditions Bacteria were obtained from the following sources: ATCC (Manassas, Va.), ARS Culture Collection (Peoria, Ill.), Dr. R. Gross (Polytechnic University, Brooklyn, N.Y.), and Dr. W. Fett (ERRC-ARS-USDA, Wyndmoor, Pa.). Cells were grown either in Luria medium (1% w/v tryptone; 0.5% w/v yeast extract; 0.5% w/v NaCl) or tryptic soy broth (Difco, Detroit, Mich.). The corresponding solid media were prepared in 1.21.5% (w/v) agar. Cells were grown at 30 C, with 250 rpm rotary shaking for the liquid cultures. Molecular biology procedures Genomic DNA was isolated by using a Wizard Genomic DNA Purication Kit (Promega, Madison, Wis.). Template DNA samples for use in colony PCR were prepared as follows: a single colony on solid growth medium was picked with a sterile toothpick into 50 ll MilliQ water in a 500-ll Eppendorf tube. The cell suspension was heated to 95 C for 10 min. Five microliters of the lysate was used in colony PCR (50 ll total volume). PCR was performed in a GeneAmp PCR System 9700 (PE Applied Biosystems, Foster City, Calif.). Forward (I-179L; 5ACAGATCAACAAGTTCTACATCTTCGAC-3) and reverse (I-179R; 5-GGTGTTGTCGTTGTTCCAGTAGAGGATGTC3) primers, custom-ordered from Life Technologies (Gaithersburg, Md.), were based on two highly conserved sequences deduced from a multiple alignment analysis of the pseudomonad phaC genes. While I-179L is also homologous to a sequence region in the type I phbC genes, the I-179R is highly specic only to the type II phaC genes. Taq DNA polymerase and ELONGASE enzyme mix (both from Life Technologies) were used following the manufacturer's protocol. PCR products were analyzed by agarose gel electrophoresis in TBE buer (0.089 M tris-base, 0.089 M boric acid, 0.002 M Na-EDTA). PCR cloning and sequencing PCR products were separated by electrophoresis on agarose gel containing 4 lg crystal violet/ml in TAE buer (50 mM tris-acetate, pH 8; 1 mM EDTA). The desired fragment was excised and eluted from the gel using the GENECLEAN II kit (BIO101, La Jolla, Calif.). The puried PCR fragment was subcloned into a pT7Blue-3 vector using a Perfectly Blunt cloning kit (Novagen, Madison, Wis.). The recombinant DNA was used to transform Escherichia coli DH5a (Life Technologies). White transformants were selected on solid Luria medium containing 100 lg ampicillin/ ml and pre-spread with both 35 ll of 5-bromo-4-chloro-3-indolylb-D-galactopyranoside (50 mg/ml) and 20 ll of 100 mM isopropylb-D-thioglucopyranoside. Plasmid DNA was puried from the

Results
PCR detection of phaC sequences We rst developed the PCR protocol using genomicDNAs puried from P. resinovorans, P. oleovorans, P. putida, P. citronellolis, and P. saccharophila. These organisms have been reported to produce mcl-PHA (Brandl et al. 1988; Cromwick et al. 1996; Eggink et al. 1993; Solaiman et al. 1999; Timm and Steinbuchel 1990). Our results showed that with the exception of P. saccharophila, a distinct 540-bp PCR product was obtained (Fig. 1). The size of the PCR product agrees with the length of the phaC1 and phaC2 genes anked by the I-179L/I-179R primer-pair. When ELONGASE, formulated by its manufacturer for long PCR amplication was used, an additional ca. 3.4-kb amplicon was

Fig. 1 Polymerase chain reaction (PCR) detection of phaC1/C2 genes. Puried genomic DNA and regular Taq DNA polymerase were used in the reactions. Lane 1 No DNA, lane 2 Pseudomonas resinovorans NRRL B-2649, lane 3 P. oleovorans NRRL B-14683, lane 4 P. putida KT2442, lane 5 P. citonellolis NRRL B-2504, lane 6 P. saccharophila NRRL B-628, lane 7 DNA size markers (k-DNA/ HindIII + /X174/HaeIII)

692

observed (data not shown). This long amplicon represents the DNA sequence anked by the I-179L and I-179R annealing sites in the phaC1 and phaC2 genes, respectively, and included the entire phaZ gene. The PCR protocol developed using the puried genomic-DNAs was then tested for the detection of phaC gene sequences in lysates prepared by heating the suspension of bacterial colonies. The results (Fig. 2) showed that the characteristic 540-bp PCR product was produced even from the crude cell-lysates of microorganisms shown to contain phaC1 and phaC2 genes (Fig. 1). Results in Fig. 2 also show that P. viridiava and P. uorescens indeed harbor the type-II pha locus. Production of PHA in these two organisms had previously been documented (Steinbuchel 1991). Interesting ly, unlike P. oleorovans NRRL B-14683, the presence of pseudomonad-type phaC genes was not detected in strain NRRL B-14682 using the present PCR protocol. These PHA synthase genes were not detected in an oildegrading Pseudomonas sp. ATCC 21909. Using this rapid and specic PCR method, several microorganisms representing pseudomonads from different rRNA homology groups were screened for the presence of the type-II pha gene locus. Data in Table 1 show that among the organisms screened, only P. corrugata 388 [belonging to the non-uorescent rRNA homology group I pseudomonad (Young et al. 1992)] contains the phaC1- and phaC2-type genes. Sequence analysis of the amplicon from P. corrugata The 540-bp PCR product from P. corrugata sample was subcloned and sequenced. A BLASTX search using the cloned 540-bp P. corrugata sequence as query produced amino acid sequence matches with more than 35 PHA

Table 1 PCR detection of type II PHA synthase genes in selected pseudomonads. PCR was performed using cell lysates prepared by boiling of the resuspension of bacterial colony as described in Materials and methods. Bacterial strains were obtained from Dr. William Fett, ERRC Bacterial strain Comamonas acidovorans ATCC 15668 C. testosteroni ATCC 11996 Pseudomonas corrugata 388 P. stutzeri ATCC 17588 P. andropogonis 27 P. andropogonis 11 Ralstonia solanacearum K60
a

rRNA-DNA Homology group III III I I II II II

phaC1/C2a ) ) + ) ) ) )

Determined as indicated by the presence or absence of 540-bp PCR amplicon

and poly-b-hydroxybutyrate (PHB) synthases (data not shown). The alignments having the most homology were those with the PHA synthase 2 (PhaC2) of Pseudomonas sp. 613 [GenBank (GB) AB014758], P. resinovorans (GB AF129396), P. putida (GB AF042276), P. oleovorans (Swiss-Prot P26496), and P. aeruginosa (Protein Information Resource S29307). Apparently, the particular bacterial clone chosen for the sequence analysis contained an amplied segment of the P. corrugata phaC2 gene. Characterization of mcl-PHA of P. corrugata When visualized by transmission electron microscopic imaging, P. corrugata 388 grown for 72 h under PHAinducing conditions exhibited the characteristic polymer-containing inclusion bodies (Fig. 3). Isolation of the polymer from the P. corrugata culture produced PHA at a crude yield of 49% cell dry weight. Composition analysis of the PHA showed that the major repeat-units were 3-hydroxyoctanoate (C8:0; 47.0 1.0 mol%), 3-hydroxydecanoate (C10:0; 24.5 0.5 mol%), and 3-hydroxytetradecenoate (C14:1; 16.5 0.5 mol%). Results of gel permeation chromatography indicated that the weight-average and number-average molecular mass of the P. corrugata PHA were 735 105 kDa and 181 45 kDa, respectively. The polydispersity of the polymer was calculated as 4.19 0.46.

Discussion
Fig. 2 PCR detection of phaC1/C2 genes in bacterial lysates. Lane 1 No bacterial lysate, lane 2 P. resinovorans NRRL B-2649, lane 3 P. oleovorans NRRL B-14682, lane 4 P. oleovorans NRRL B-14683, lane 5 P. putida KT2442, lane 6 P. putida ATCC 17391, lane 7 P. citronellolis NRRL B-2504, lane 8 P. viridiava ATCC 13223, lane 9 P. uorescens ATCC 17824, lane 10 Pseudomonas sp. ATCC 21909, lane 11 P. saccharophila NRRL B-628, lane M DNA size markers

The PCR screening protocol described in this paper is a rapid, simple, and specic method for identifying mclPHA-producing microorganisms. The method is especially important for the identication and verication of organisms that harbor mcl-PHA biosynthesis genes. Lipophilic dye-based screening procedures (Spiekermann et al. 1999; Takagi and Yamane 1997) are useful and

693

Fig. 3 Thin-section electron microscopic images of P. corrugata 388. Cells were cultured in chemically dened E* medium supplemented with glucose and oleic acid. Incubation was at 30C for 72 h with 250 rpm rotary shaking

direct for visualizing the presence of PHA inclusion bodies in microbes. However, these methods cannot dierentiate between scl-PHA, mcl-PHA, and other lipid materials such as triacylglycerols and waxes. Furthermore, since PHA biosynthesis often depends on specic growth conditions (Anderson and Dawes 1990), the staining methods could overlook potential PHA producers that do not contain the polymer at the time of screening. Moreover, in contrast to elaborate and timeconsuming hybridization methods (Schembri et al. 1994; Timm et al. 1994), this PCR procedure also provides a means for quick, simple, and mcl-PHA-specic detection of the type II PHA biosynthesis genes, using only cell lysate as the source of DNA template. Numerous microorganisms have been reported to synthesize PHAs. In many cases, the type of PHA produced was not clearly dened. For example, P. corrugata is characterized as a PHB producer based on Nile blue A staining (Sutra et al. 1997). However, the repeatunit composition of the polymer had not previously been characterized. In this study, we conclusively showed that P. corrugata contains the type II PHA biosynthesis gene locus and proceeded to conrm the production of mclPHA by this organism. Using a similar approach, other organisms previously reported as PHA producers can now be veried with respect to the type of their PHA biosynthesis genes and presumably the class of polymer

itself. The simultaneous existence of type I PHA genes cannot be ruled out by the present PCR procedure. However, in conjunction with broad-specicity DNAhybridization methods (Schembri et al. 1994; Timm et al. 1994), this PCR procedure can serve as a powerful tool for characterizing PHA-biosynthesis gene loci in microorganisms. Randomly selected pseudomonads from the three rRNA-DNA homology groups were screened for the presence of type II PHA synthase genes (Table 1). The results showed that only P. corrugata harbors the phaC1/C2 genes, further supporting the observation that mcl-PHAs are mostly produced by Pseudomonas belonging to the rRNA-DNA homology group I (Steinbuchel 1991). We also observed that strain varia tion is important in determining the PHA-producing potential of a bacterial species. For example, while phaC1/C2 genes were detected (Table 1) in the mclPHA-producing P. oleovorans NRRL B-14683 (ATCC 29347;Tf41l, Schwartz), these type II PHA synthase genes were not found in P. oleovorans NRRL B-14682 (ATCC 13474). A separate experiment conrmed that mcl-PHA was not synthesized by strain NRRL B-14682 (data not shown). Similarly, our results showed that P. stutzeri ATCC 17588 did not harbor the phaC1/C2 genes (Table 1). He et al. (1998), however, reported that P. stutzeri 1317 was an mcl-PHA producer. Some of the organisms listed in Table 1 (e.g., Comamonas acidovorans, C. testosteroni, and R. solanacearum) have been reported as PHA producers (Steinbuchel 1991). The failure to detect the specic PCR amplicon in these microorganisms may be indicative of strain variation as described above. Alternatively, these organisms may contain non-type II PHA synthase genes that would not be detected by the present PCR procedure. A multiplex PCR method that combines the current phaC1/C2-specic primer-pair with non-type II pha-specic PCR oligomers should allow for simultaneous detection or verication of all classes of PHA synthase genes. An important discovery in this study is that P. corrugata contains phaC1/C2 genes and produces a mcl-PHA. Even though this species is classied as a polyhydroxyalkanoate-accumulating microbe, it was long assumed to synthesize PHB (Sutra et al. 1997). Results from this study, however, conclusively show that a mcl-PHA is produced by this bacterium. More importantly, the study showed that the PHA polymer produced by P. corrugata has the highest molecular weight value among the unsaturated mcl-PHAs reported to date (Ashby and Foglia 1998; Brandl et al. 1988). Accordingly, this bacterial species may be an important organism for the production of high molecular weight mcl-PHAs.
Acknowledgements We thank Nicole Cross, Rob DiCiccio, and Laurie Fortis for technical assistance; Dr. Peter H. Cooke and Douglas Soroka (Microscopic Imaging Group, Core Technologies Unit, ERRC) for acquiring the electron microscopic images of P. corrugata; and Dr. Alberto Nunez for performing GC/MS analyses.

694

References
Altschul SF, Madden TL, Schaer AA, Zhang J, Zhang Z, Miller W, Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 33893402 Anderson AJ, Dawes EA (1990) Occurrence, metabolism, metabolic role, and industrial uses of bacterial polyhydroxyalkanoates. Microbiol Rev 54: 450472 Ashby RD, Foglia TA (1998) Poly(hydroxyalkanoate) biosynthesis from triglyceride substrates. Appl Microbiol Biotechnol 49: 431437 Brandl H, Gross RA, Lenz RW, Fuller RC (1988) Pseudomonas oleovorans as a source of poly(b-hydroxyalkanoates) for potential applications as biodegradable polyesters. Appl Environ Microbiol 54: 19771982 Cromwick A-M, Foglia T, Lenz RW (1996) The microbial production of poly(hydroxyalkanoates) from tallow. Appl Microbiol Biotechnol 46: 464469 Eggink G, van der Wal H, Huijberts GNM, de Waard P (1993) Oleic acid as a substrate for poly-3-hydroxyalkanoate formation in Alcaligenes eutrophus and Pseudomonas putida. Indian Crops Products 1: 157163 He W, Tian W, Zhang G, Chen G-Q, Zhang Z (1998) Production of novel polyhydroxyalkanoates by Pseudomonas stutzeri 1317 from glucose and soybean oil. FEMS Microbiol Lett 169: 4549 Lee SY (1996) Bacterial polyhydroxyalkanoates. Biotechnol Bioeng 49: 114 McCool GJ, Cannon MC (1999) Polyhydroxyalkanoate inclusion body-associated proteins and coding region in Bacillus megaterium. J Bacteriol 181: 585592 Poirier Y, Nawrath C, Somerville C (1995) Production of polyhydroxyalkanoates, a family of biodegradable plastics and elastomers, in bacteria and plants. Biotechnology 13: 142150 Schembri MA, Bayly RC, Davies JK (1994) Cloning and analysis of the polyhydroxyalkanoic acid synthase gene from an Acinetobacter sp.: evidence that the gene is both plasmid and chromosomally located. FEMS Microbiol Lett 118: 145152

Solaiman, DKY, Ashby, RD, Foglia TA (1999) Medium-chainlength poly(b-hydroxyalkanoate) synthesis from triacylglycerols by Pseudomonas saccharophila. Curr Microbiol 38: 151154 Spiekermann P, Rehm BHA, Kalscheuer R, Baumeister D, Steinbuchel A (1999) A sensitive, viable-colony staining method using Nile Red for direct screening of bacteria that accumulate polyhydroxyalkanoic acids and other lipid storage compounds. Arch Microbiol 171: 7380 Steinbuchel A (1991) Polyhydroxyalkanoic acids. In: Byrom D (ed) Biomaterials: novel materials from biological sources. Stockton Press, New York, pp 123213 Steinbuchel A, Fuchtenbusch B (1998) Bacterial and other bio logical systems for polyester production. Trends Biotechnol 16: 419427 Sutra L, Siverio F, Lopez MM, Hunault G, Bollet C, Gardan L (1997) Taxonomy of Pseudomonas strains isolated from tomato pith necrosis: emended description of Pseudomonas corrugata and proposal of three unnamed uorescent Pseudomonas genomospecies. Int J Syst Bacteriol 47: 10201033 Takagi Y, Yamane T (1997) Replica technique for screening poly(3-hydroxyalkanoic acid)-producing bacteria by Nile blue staining. J Ferment Bioeng 83: 121123 Tatiana AT, Madden TL (1999) Blast 2 sequences a new tool for comparing protein and nucleotide sequences. FEMS Microbiol Lett 174: 247250 Timm A, Steinbuchel A (1990) Formation of polyesters consisting of medium-chain-length 3-hydroxyalkanoic acids from gluconate by Pseudomonas aeruginosa and other uorescent pseudomonads. Appl Environ Microbiol 56: 33603367 Timm A, Wiese S, Steinbuchel A (1994) A general method for identication of polyhydroxyalkanoic acid synthase genes from pseudomonads belonging to the rRNA homology group I. Appl Microbiol Biotechnol 40: 669675 Young JM, Takikawa Y, Gardan L, Stead DE (1992) Changing concepts in the taxonomy of plant pathogenic bacteria. Annu Rev Phytopathol 30: 67105