Molecular Characterization of the phaEC_Hm Genes, Required for Biosynthesis of Poly(3-Hydroxybutyrate) in the Extremely Halophilic Archaeon Haloarcula marismortui

Abstract

Although many haloarchaea produce biodegradable polyhydroxyalkanoates (PHAs), the genes involved in PHA synthesis in the domain of Archaea have not yet been experimentally investigated yet. In this study, we revealed that Haloarcula marismortui was able to accumulate poly(3-hydroxybutyrate) (PHB) up to 21% of cellular dry weight when cultured in a minimal medium with excessive glucose and identified the phaE_Hm and phaC_Hm genes, probably encoding two subunits of a class III PHA synthase. These two genes were adjacent and directed by a single promoter located 26 bp upstream of the transcriptional start site and were constitutively expressed under both nutrient-rich and -limited conditions. Interestingly, PhaC_Hm was revealed to be strongly bound with the PHB granules, but PhaE_Hm seemed not to be. Introduction of either the phaE_Hm or phaC_Hm gene into Haloarcula hispanica, which harbors highly homologous phaEC_Hh genes, could enhance the PHB synthesis in the recombinant strains, while coexpression of the both genes always generated the highest PHB yield. Significantly, knockout of the phaEC_Hh genes in H. hispanica led to a complete loss of the PHA synthase activity. Complementation with phaEC_Hm genes, but not a single one, restored the capability of PHB accumulation as well as the PHA synthase activity in this phaEC-deleted haloarchaeon. These results indicated that the phaEC genes are required for biosynthesis of PHB and might encode an active PHA synthase in the Haloarcula species.

Polyhydroxyalkanoates (PHAs) are a class of biodegradable polyesters of (R)-hydroxyalkanoates. These water-insoluble biopolymers are accumulated by a wide variety of bacteria and haloarchaea when the carbon source is available in excess but other nutrients are growth limiting (29). PHAs are attracting increasing attention due to their biodegradable, biocompatible, and thermoplastic features; thus, they are potential substitutes for petrochemical-derived plastics and can be used as packaging and biomedical materials, nonwoven fabrics, and flavor delivery agents (26). Nevertheless, the expensive cost for PHA production has limited its large-scale application. In this regard, the haloarchaeal strains employed as PHA producers have many advantages (17, 23). First, haloarchaea are a group of extreme archaea growing optimally in hypersaline environments (25). As there are very few microorganisms that are able to survive and grow at high salinities, the risks of microbial contamination can be reduced. Second, some cheaper carbon sources, such as starch, could be used to synthesize PHAs by haloarchaea, and so the cost of PHA production would be lowered. Third, haloarchaea can be easily lysed in distilled water, which could avoid the use of large quantities of organic solvent and save time for PHA preparation. Moreover, although high concentrations of saline minerals in the culture medium may cause corrosion of fermentors, a novel corrosion-resistant bioreactor has been reported (10). Therefore, PHA production in haloarchaea will have much economic potential after we have a thorough understanding of PHA synthesis in these halophilic archaea.

The presence of PHA granules in haloarchaea was first reported in 1972 (15). Since then, many haloarchaeal genera, including Haloferax, Halobacterium, Haloarcula, and Haloquadratum, have been found to accumulate kinds of PHAs, such as poly(3-hydroxybutyrate) (PHB) or poly(3-hydroxybutyrate-co-hydroxyvalerate) (6, 10, 15, 17, 19). Some of these halophilic archaea, including haloarchaeal strain 56 and Haloferax mediterranei ATCC 33500, have been shown to accumulate large amounts of PHA (∼60% of cellular dry weight) under optimized culture conditions (6, 10, 23). However, although the enzymatic properties of PHA synthase from the crude extracts of the halophilic archaeon strain 56 have been investigated (11), the molecular features of the PHA synthase in the domain of Archaea are still far from well characterized. PHA synthases are the central enzymes during the polymerization of 3-hydroxyacyl-coenzyme A (CoA) thioesters into PHAs. The characteristics of PHA synthases are closely related to the monomer composition, molecular weight, and the amount and polymer properties of the generated PHAs (14, 34). At present, PHA synthases from bacteria are grouped into four classes according to their substrate specificity and the subunit compositions (36). The PHA synthase of Allochromatium vinosum (originally named Chromatium vinosum), consisting of two subunits, PhaE_Av (40.53 kDa) and PhaC_Av (39.73 kDa) (22), is a representative of the class III synthases. Many comprehensive studies of this bacterial PHA synthase have been performed in vivo and in vitro (13, 24, 27, 38, 41). In haloarchaea, the genes encoding the homologues of class III PhaC were already annotated in the recently published genome sequences of Haloarcula marismortui ATCC 43049 (1) and Haloquadratum walsbyi DSM 16790 (3). This has provided an opportunity to investigate the PHA synthesis pathways in the haloarchaea.

In this study, we report the molecular characterization of the phaEC_Hm operon that is required for PHB synthesis in the extremely halophilic archaeon H. marismortui. The expression profile of the phaEC_Hm genes was analyzed, and the significance of the two genes in PHB synthesis was identified through both heterologous expression and gene knockout/complementation investigations. We suggest that both PhaC_Hm and PhaE_Hm are necessary for an active PHA synthase in H. marismortui.

MATERIALS AND METHODS

Strains and growth conditions.

The strains used in this study are listed in Table 1. Escherichia coli was grown in Luria-Bertani medium at 37°C (30). Unless otherwise noted, H. marismortui, H. hispanica, and Haloferax volcanii were cultivated at 37°C in nutrient-rich medium AS-168 (per liter, 5.0 g Bacto Casamino Acids, 5.0 g yeast extract, 1.0 g sodium glutamate, 3.0 g trisodium citrate, 200 g NaCl, 20 g MgSO₄ · 7H₂O, 2.0 g KCl, 0.36 g FeSO₄ · 4H₂O, and 0.36 mg MnCl₂ · 4H₂O, pH 7.2). For analysis of PHA accumulation in haloarchaea, nutrient-limited minimal medium (per liter, 200 g NaCl, 20 g MgSO₄ · 7H₂O, 2 g KCl, 1 g sodium glutamate, 37.5 mg KH₂PO₄, 50 mg FeSO₄ · 7H₂O, 0.36 mg MnCl₂ · 4H₂O, 1 g yeast extract, pH 7.2) supplemented with 2% glucose, resulting in the PHA accumulation medium MG, was explored. Cells were first cultivated in AS-168 medium for 72 h and then transferred into MG medium and continuously cultured for 48 h on a rotary shaker. Afterwards, 100 ml of these seed cultures was inoculated into 2 liters of MG medium for fermentation in a 5-liter computer-controlled fermentor (NBS Bioflo 110) and cultured for 4 to ∼10 days with respect to H. hispanica or H. marismortui strains. When needed, 3 μg/ml mevinolin, 100 μg/ml ampicillin, or 50 μg/ml kanamycin was added into the medium.

TABLE 1.

Strains and plasmids used in this study

Strain or plasmid	Relevant characteristics	Source or reference
Strains
Haloarcula marismortui ATCC 43049	Wild-type strain	1
Haloarcula hispanica ATCC 33960	Wild-type strain	ATCC
Haloarcula hispanica PHB-1	phaEC-deleted mutant of H. hispanica	This study
Haloferax volcanii DS70	Wild-type strain cured of endogenous pHV2 plasmid	40
Escherichia coli JM109	recA1 supE44 endA1 hsdR17 gyrA96 relA1 thi	30
Escherichia coli BL21(DE3)	F−ompT hsdSB(rB− mB−) dcm gal (DE3)	Novagen
Plasmids
pWL102	10.5-kb shuttle vector, Ampr Mevr	18
pWLE	11.2 kb, phaE and its native promoter	This study
pWLfdxC	12.1 kb, promoter of fdx and phaC	This study
pWLEC	12.6 kb, phaEC and its native promoter	This study
pET28a	5.4 kb, IPTG-inducible expression vector with His tag	Novagen
pET28aE	Expression plasmid for His tag fused to PhaE	This study
pET28aC	Expression plasmid for His tag fused to PhaC	This study
pUBP2	12.3-kb shuttle vector, Ampr Mevr	2
pUBP	6.6 kb, derivative of pUBP2 created by removing the pHH9-ori region	This study
pUBPHL	7.6 kb, integration vector for knockout of phaECHh of H. hispanica ATCC 33960	This study

Plasmid construction, transformation, and gene knockout.

The plasmids and primers used in this study are listed in Tables 1 and 2, respectively. The expression plasmids for haloarchaea were derived from pWL102, a shuttle vector between E. coli and haloarchaea (Table 1). Unless otherwise noted, the inserts were PCR amplified from the genomic DNA of H. marismortui ATCC 43049. Plasmids pWLE and pWLEC were constructed by, respectively, cloning of phaE_Hm (amplified with primers phaEF1 and phaER1) and phaEC_Hm (amplified with primers phaEF1 and phaCR1) into pWL102, with the native promoter of phaEC_Hm. For overexpression of the phaC_Hm gene, the promoter sequence of the fdx_Hm gene (amplified with primers fdxF and fdxR) and coding sequence of phaC_Hm (amplified with primers phaCF1 and phaCR1) were inserted into pWL102, resulting in the plasmid pWLfdxC. For expression of PhaE_Hm and PhaC_Hm in E. coli BL21(DE3), phaE_Hm and phaC_Hm genes were amplified by primer pairs phaEF2/phaER2 and phaCF2/phaCR2 and cloned into the expression vector pET28a, resulting in pET28aE and pET28aC, respectively (Table 1).

TABLE 2.

Primers used in this study

Primer	Sequence (5′ to 3′)a
fdxF	ATACCATGGGTCGACGGCCGGCAGCAC
fdxR	GGCGGATCCCGGCATCACTGCAGAGTT
phaCF1	CTAGGATCCATGTCCAGCAACCCCTTC
phaCR1	CGTGGTACCTTACAGTTGATCGAGCCA
phaCF2	TTAGGATCCTCCAGCAACCCCTTCAAT
phaCR2	GCGGAATTCTTACAGTTGATCGAGCCA
phaEF1	ATTGGATCCCAGCTCGAAGAAGTGCAG
phaER1	GGCGGTACCTTATTCTTCTAAGTGTTC
phaEF2	GGCGGATCCATGAGTAATACAAACAAC
phaER2	GCGGAATTCTTATTCTTCTAAGTGTTC
RTEF	ACATGAAGGCCCAGGCGG
RTER	CGTCCGCTTCCTGTTGCA
RTECF	GGAAGCGGACGACCTGAG
RTECR	CGGTTGATGAGCGCGTAA
RTCF	GTTTACGCGCTCATCAACCGACCG
RTCR	TCGAGGACGCCGCCAGTGTG
TIP	CTCCTCAACCATCTCCGTCC
SeqR	CACGACTCCACGAAGGCCGC
CrossF1	GCGAAGCTTCCAGCAAAACGTCAACAG
CrossR1	TATGGATCCCTTCGGCCTGCACTTCCT
CrossF2	TATGGATCCCTCGAAGACGTGTATCAGGAC
CrossR2	TATGGTACCCGGGATTCGGTGGTTTCG

^aSequences representing restriction sites are underlined.

For knockout of phaEC_Hh genes in H. hispanica ATCC 33960, a 434-bp DNA fragment located immediately upstream of the phaEC_Hh operon and a 563-bp fragment at the 3′ region of the phaC_Hh gene were amplified by primer pairs CrossF1/CrossR1 and CrossF2/CrossR2, respectively (Table 2). These two DNA fragments were linked and inserted into the plasmid pUBP, a derivative of pUBP2 with the pHH9-ori region omitted by EcoRI digestion, resulting in the integration plasmid pUBPHL (Table 1). Transformation of pUBPHL into ATCC 33960 and screening of a phaEC_Hh deletion mutant, named H. hispanica PHB-1, were carried out according to the method of Tu et al. (39). The plasmids pWLE, pWLfdxC, and pWLEC were transformed into H. hispanica PHB-1 to verify the functions of phaEC_Hm genes.

These haloarchaeal plasmids were usually constructed in E. coli JM109 at first and then transformed into H. volcanii and H. hispanica with the polyethylene glycol-mediated transformation method described by Cline et al. (4). The PCR-amplified sequence in each construct was verified by DNA sequencing.

Isolation and characterization of PHA granules.

Haloarchaeal cells accumulating PHA were harvested by centrifugation and washed once with basal salt solution (BSS; per liter, 200 g NaCl, 20 g MgSO₄ · 7H₂O, 2 g KCl, and 3 g trisodium citrate). Cellular PHA content and its composition were analyzed by gas chromatography (GC). Briefly, the lyophilized cells were subjected to methanolysis in a mixture of chloroform and methanol containing 3% (vol/vol) sulfuric acid at 100°C for 4 h. The resulting hydroxyacyl methylesters were then analyzed with a GC-6820 apparatus (Agilent). The PHB content (as percent weight/weight) in the cells was calculated as follows: (mass of PHB/original lyophilized cell mass) × 100%.

For isolation of PHA granules, PHA-accumulating cells were centrifuged, washed once with BSS, and resuspended in a solution containing 20 mM Tris-HCl (pH 7.5), 200 g/liter NaCl, 20 g/liter KCl, and 5 g/liter MgSO₄ · 7H₂O. Crude extracts were obtained by passing cells three times through a French press (13,000 lb/in²; Thermo IEC). Intact cells and debris were pelleted by centrifugation (10,000 × g, 4°C, 30 min). PHA granules in the supernatants were collected by ultracentrifugation (200,000 × g, 4°C, 2 h) and then suspended in the same buffer and stored at −70°C for future use.

TEM.

H. marismortui ATCC 43049 cells cultivated in AS-168, MG, or MGC medium (MG medium plus 0.5% Casamino Acids) for 72 h were harvested by centrifugation, washed twice with BSS, and then subjected for transmission electron microscopic (TEM) analysis as described by Pötter et al. (28). Micrographs were recorded with a Hitachi H-700A electron microscopy.

Reverse transcriptase PCR and primer extension.

Total RNA of H. marismortui ATCC 43049 was isolated with the TRIzol reagent (Gibco BRL) from cells cultured for 72 h. Prior to the reverse transcription (RT) reaction, about 10 μg of total RNA was treated with RNase-free RQ1 DNase (Promega) to eliminate any contamination of DNA. The DNA-free RNA was used for cDNA synthesis with SuperScript III reverse transcriptase (Invitrogen) using the antisense primer RTCR (Table 2). After incubation for 1 h at 42°C, the RT reaction was terminated by incubation at 70°C for 20 min, and 1 μl of reverse transcripts was applied for PCR with the primer pair RTEF/RTER for the phaE_Hm gene, RTCF/RTCR for phaC_Hm, and RTECF/RTECR for phaEC_Hm (Table 2). These RT-PCR products were separated in 1% agarose gel, followed by ethidium bromide staining.

For determination of the transcriptional start site of phaEC_Hm genes, total RNA was obtained from the H. volcanii transformants harboring pWLEC or pWL102. The RNA samples (6 μg) were used as templates for primer extensions, while the PCR product amplified with primer pair phaEF1/seqR was used as template for DNA sequencing. The primer TIP (Table 1) was labeled at the 5′ end with [γ-³²P]ATP and was used for both DNA sequencing and primer extension as described previously (42).

Expression and purification of PhaE_Hm-His₆ and PhaC_Hm-His₆ in E. coli.

Plasmids of pET28aE and pET28aC were constructed to express PhaE_Hm-His₆ and PhaC_Hm-His₆ in E. coli BL21(DE3). The E. coli recombinants were cultured until the mid-exponential phase and then induced with isopropyl-β-d-thiogalactopyranoside (IPTG; 1 mM) for an additional 4 h. The His-tagged proteins were purified using the His-bind column (Novagen) according to the manufacturer's instructions. The concentrations of proteins were determined with the bicinchoninic acid protein assay kit (Pierce).

Antisera preparation and Western blot analysis.

The antisera were prepared by immunizing mice with purified PhaE_Hm-His₆ and PhaC_Hm-His₆ as previously described (37). For Western blot analysis, crude extracts from H. marismortui ATCC 43049 were obtained by disrupting cells through a French press as described above. Soluble fractions were obtained by centrifugation at 10,000 × g for 30 min. One hundred micrograms of proteins from cellular extracts and PHA granules were separated on 12% sodium dodecyl sulfate-polyacrylamide gels, and 0.5-μg aliquots of purified PhaE_Hm-His₆ and PhaC_Hm-His₆ proteins were used as controls. The proteins on the gel were then transferred onto a polyvinylidene difluoride membrane by using a semidry electrophoretic transfer cell (Bio-Rad) at 15 V for 15 min. The blots were blocked in 7% nonfat dry milk in phosphate-buffered saline (pH 7.4) for 30 min and then incubated with the anti-PhaE_Hm-His₆ or anti-PhaC_Hm-His₆ antiserum for 3 h. After appropriate washing, the blots were incubated with alkaline phosphatase-labeled goat anti-mouse immunoglobulin G for 30 min at room temperature, followed by staining in the alkaline phosphatase development solution containing nitroblue tetrazolium and 5-bromo-4-chloro-3-indolylphosphate.

PHA synthase activity assay.

PHA synthase activity was measured spectrophotometrically by recording the release of CoA during the polymerization of 3-hydroxybutyryl-CoA (3HB-CoA). The assay mixture (1 ml) contained 20 mM Tris-HCl (pH 7.5), 3.4 M KCl, 100 μM Mg(CH₃COO)₂, 100 μM 3HB-CoA, 1 g/liter bovine serum albumin, and 500 μg of protein extracts. The reaction was initiated by the addition of PHA synthase samples at 30°C. The absorbance at 412 nm was measured at defined time points. The concentration of CoA was determined using a molar absorption coefficient of 13,600 M⁻¹ cm⁻¹ (5). One unit was defined as the amount of enzyme that catalyzed the generation of 1 μmol CoA per min.

Nucleotide sequence accession number.

According to the DNA sequence of phaEC_Hm of H. marismortui ATCC 43049 (GenBank accession no. AY596297), the highly homologous phaEC_Hh genes of H. hispanica ATCC 33960 were cloned and sequenced and deposited in GenBank under the accession number EU022705.

RESULTS

PHA accumulation in H. marismortui under nutrient-limited conditions.

The capability for PHA synthesis in H. marismortui ATCC 43049 was investigated under different culture conditions (see Materials and Methods). Although almost no PHA synthesis was observed when this haloarchaeal strain was cultivated in nutrient-rich medium AS-168 (Fig. 1A) or minimal medium with Casamino Acids and glucose (MGC medium, pH 7.2) (Fig. 1B), large amounts of PHA-insoluble granules were detected in the cells when cultured in minimal medium with 2% glucose as the excessive carbon source (MG medium, pH 7.2) (Fig. 1C). The typical PHA granules contributed a significant fraction of the cell volume, with similar shape and size (0.1 to 0.5 μm in diameter) as those in bacteria. These observations were confirmed by GC analysis, with which the PHA granules were further identified to be PHB.

FIG. 1.

TEM images of H. marismortui ATCC 43049 cells grown at 37°C for 72 h in different media: (A) nutrient-rich AS-168 medium; (B) minimal medium supplemented with 2% glucose and 0.5% Casamino Acids (MGC medium); and (C) minimal medium supplemented with 2% glucose (MG medium). Bar, 0.5 μm.

To investigate the relationship between cell growth and PHB content in H. marismortui ATCC 43049 when cultured in MG medium, the time profile of the PHB accumulation was determined (Fig. 2). The PHB content increased quickly in the first 72 h and remained relative stable, at about 17% (wt/wt) of cellular dry weight from 72 h to 168 h. The PHB content reached its maximum at 21% (wt/wt) after 180 h of fermentation and then began to decrease slowly (Fig. 2). This was a typical result of several independent investigations. Thus, in the following experiments when we analyzed gene expression and PHA synthase activity as well as PHB accumulation, we usually collected the cells at 72 to ∼180 h of fermentation, corresponding to an atypical exponential phase. It is notable that some other factors, including concentration of nitrogen and phosphorus as well as the oxygen limitation that might affect the PHB accumulation, were not investigated in this study.

FIG. 2.

Profiles of PHB accumulation and cell growth of H. marismortui ATCC 43049 cultivated in MG medium in a 5-liter fermentor (NBS Bioflo 110) at 37°C for about 10 days. The PHB content and cell density, as the optical density at 600 nm (OD₆₀₀), were monitored. This figure represents a typical result of three independent experiments.

Identification of phaEC_Hm genes for PHB synthesis in H. marismortui.

The genome sequence of H. marismortui ATCC 43049 has been published recently (1). In this genome there is an open reading frame (ORF) annotated as the phaC_Hm gene, which encodes a 475-amino-acid protein. PhaC_Hm contains a highly conserved “lipase box-like” sequence (Gly-X-Cys-X-Gly-Gly) (Fig. 3A), which is believed to be an active site of PHA synthase (7). Moreover, the amino acids Cys162-Asp317-His346, as equivalents of the catalytic triad residues Cys149-Asp303-His331 in PhaC_Av (12, 27), were also detected in this protein. Significantly, PhaC_Hm showed high homology (37 to 43% identical amino acids) with the PhaC subunits of the bacterial class III PHA synthases and contained the conserved motif Arg-Met-Glu-X-Trp-Ile-X-Asp-X-X-Asp (Fig. 3A) that is typically found in class III synthases (9). These results suggested that PhaC_Hm might be one subunit of the class III PHA synthase in H. marismortui. Interestingly, immediately upstream of phaC_Hm gene, we detected another ORF with unknown function (Fig. 4A) which encoded a protein showing sequence similarity (21 to 25% identical amino acids) with the PhaE subunits of class III PHA synthases in bacteria (8) and contained a typical “PhaE box” (Fig. 3B). Therefore, we designated this ORF as phaE_Hm gene and propose that the phaEC_Hm genes encode two subunits of a functional PHA synthase in H. marismortui.

FIG. 3.

Partial alignments of the amino acid sequences of PhaC (A) and PhaE (B) subunits from H. marismortui ATCC 43049 (Hm), H. walsbyi DSM 16790 (Hw), Synechococcus sp. strain MA19 (MA19), Ectothiorhodospira shaposhnikovii (Es), Allochromatium vinosum (Av), and Thiocystis violacea (Tv). Amino acids are given in standard one-letter abbreviations, and the numbers indicate the positions of the amino acids within the respective proteins. The conserved residues are darkly shaded, and the residues identical in five of the six (for PhaC) or three of the four (for PhaE) are lightly shaded. The lipase-like box, highly conserved motif of class III synthase, and PhaE box are indicated. The conserved catalytic triad residues are shown with asterisks. GenBank accession numbers are as follows: PhaC_Hm, YP_137339; PhaC_Hw, YP_658052; PhaC_MA19, AAK38139; PhaC_Es, AAG30259; PhaC_Av, S29274; PhaC_Tv, AAC60430; PhaE_Hm, YP_137338; PhaE_Es, AAG30260; PhaE_Av, P45372; PhaE_Tv, AAC60429. The corresponding PhaE sequences were not identified in H. walsbyi DSM 16790 and Synechococcus sp. strain MA19.

FIG. 4.

Transcription analysis of phaEC_Hm genes. (A) Structure and organization of the phaEC_Hm genes in H. marismortui. The putative promoter containing a TATA box (double underlined) and a transcription factor B recognition element (single underlined) and the translational start codon of phaE_Hm (boxed) are indicated. F1, F2, and F3 represent the locations of the RT-PCR products (shown in panel C) in the phaEC_Hm genes. (B) Mapping of the transcriptional start site of phaEC_Hm by primer extension. The relevant sequence is shown on the right. The transcriptional start site is indicated with an arrow. Lane 1, H. volcanii/pWL102, negative control; lane 2, H. volcanii/pWLEC; lanes CTAG, standard sequencing reaction mixture to size the mapping signals. (C) RT-PCR products of partial phaE_Hm (F1; 314 bp), phaEC_Hm (F2; 434 bp), and phaC_Hm (F3; 341 bp) sequences. The following templates were used in the PCR: lane 1, total RNA without reverse transcriptase; lane 2, H. marismortui genomic DNA; lane 3, reverse transcripts of total RNA isolated from cells grown in AS-168 medium; lane 4, reverse transcripts of total RNA isolated from cells grown in MG medium; lane M, DNA marker.

Cotranscription of phaEC_Hm genes from the same promoter.

By analysis of the upstream sequence of the phaEC_Hm genes, we detected a putative archaeal TATA box (TTTTTTT) and a transcription factor B recognition element (GG), which might be the promoter controlling the transcription of both phaE_Hm and phaC_Hm, as there was only a 2-base spacing between these two genes (Fig. 4A). Primer extension analysis revealed that the transcription start point of the phaE_Hm gene was located 26 bp downstream of this putative promoter, exactly at base A of the start codon (AUG) of the phaE_Hm gene (Fig. 4B). Thus, it was another example of a “leaderless” transcript in haloarchaea (31).

To determine whether phaE_Hm and phaC_Hm were cotranscribed under the same promoter, RT-PCR was performed to check the existence of a polycistronic transcript of phaEC_Hm. Indeed, the PCR products (F1, F2, and F3) corresponding to the partial sequences of phaE_Hm, phaC_Hm, and phaEC_Hm mRNA with expected lengths were generated by using the primer pairs RTEF/RTER, RTCF/RTCR, and RTECF/RTECR (Table 1), respectively (Fig. 4C). These results indicated that phaE_Hm and phaC_Hm were cotranscribed and were constitutively expressed no matter whether under the PHA accumulation condition (e.g., in MG medium) or not (e.g., in AS-168 medium), which was also confirmed by Northern blot analysis (data not shown).

Western blot analysis of PhaE_Hm and PhaC_Hm proteins.

To reveal the levels of PhaE_Hm and PhaC_Hm and their relationship with PHB synthesis in H. marismortui ATCC 43049, Western blotting was performed to detect the production of PhaC_Hm and PhaE_Hm under different culture conditions. Both PhaC_Hm and PhaE_Hm were detected in the cellular extracts (Fig. 5), but interestingly, while a much stronger signal of PhaC_Hm was detected from the PHA granules (Fig. 5B), no signal of PhaE_Hm was detected (Fig. 5A). This suggested that PhaC_Hm was stably attached to the PHA granules while PhaE_Hm might not or just weakly combine with the PHA granules. Thus, PhaE_Hm and PhaC_Hm likely constitute, if anything, a novel form of class III PHA synthases in H. marismortui.

FIG. 5.

Western blot analysis of cellular extracts and PHA granules from H. marismortui ATCC 43049 with antiserum against PhaE_Hm (A) or PhaC_Hm (B). Lane 1, crude extracts from cells grown in AS-168 medium; lane 2, crude extracts from cells grown in MG medium; lane 3, proteins from isolated PHA granules. The purified PhaE_Hm-His₆ (CK1) and PhaC_Hm-His₆ (CK2) from E. coli were used as controls. H. marismortui ATCC 43049 was cultivated at 37°C for 72 h.

In accordance with the RT-PCR results (Fig. 4C), both proteins (PhaE_Hm and PhaC_Hm) in the crude extracts of cells that were cultivated in either AS-168 or MG medium showed a similar amount in Western blot analysis (Fig. 5), suggesting that PhaE_Hm and PhaC_Hm are indeed constitutively expressed. We also found that the PHA synthase activity of H. marismortui ATCC 43049 was similar when grown in both media, indicating that the activity of the PHA synthase was also not regulated at the protein level (data not shown). This constitutive expression of the putative PHA synthase was different from that of the halophilic archaeon strain 56 (11) but has been widely observed in bacteria (7, 16, 33).

Overexpression of PhaE_Hm and PhaC_Hm in H. hispanica.

To demonstrate the function of phaEC_Hm genes, the DNA fragments containing phaE_Hm, phaC_Hm, and phaEC_Hm were inserted into pWL102, resulting in pWLE, pWLfdxC, and pWLEC, respectively (see Materials and Methods). These plasmids were introduced into H. hispanica ATCC 33960, a haloarchaeon that is phylogenetically close to H. marismortui ATCC 43049 and is much more easily transformed.

H. hispanica ATCC 33960 harbors a highly homologous phaEC_Hh operon (GenBank accession no. EU022705), which exhibits more than 92% identity with phaEC_Hm. As expected, overexpression of phaE_Hm and/or phaC_Hm in ATCC 33960 promoted PHB accumulation in all of the recombinant cells, e.g., at 144 h, the PHB content in the cells could be increased from 9.9% (wt/wt) to up to 16.0 to 18.0% (wt/wt) (Table 3). Interestingly, overexpression of phaE_Hm or phaC_Hm did not affect the biomass, while coexpression of phaEC_Hm genes slightly promoted the cellular dry weight, which was increased from 12.4 g/liter (for 33960/pWL102) to 15.4 g/liter (for 33960/pWLEC) after 144 h of fermentation (Table 3). Thus, the PHB yields in 33960/pWLE and 33960/pWLfdxC were, respectively, increased up to 167% (∼2.0 g/liter) and 175% (∼2.1 g/liter) of that in vector-transformed cells (∼1.2 g/liter) and, significantly, coexpression of phaEC_Hm had the most significant effect on PHB synthesis. The PHB yield in 33960/pWLEC increased up to 233% (∼2.8 g/liter) of that in the control cells after 144 h of fermentation (Table 3). These results suggested that both phaE and phaC genes were involved in PHA synthesis in the Haloarcula species.

TABLE 3.

PHB accumulation in H. hispanica ATCC 33960 recombinant strains

Strain	PHB accumulation after:
	96 h		144 h
	CDW (g/liter)	PHB/CDW (% [wt/wt])	CDW (g/liter)	PHB/CDW (% [wt/wt])
33960/pWL102	11.2	8.3	12.4	9.9
33960/pWLE	11.5	11.3	12.6	15.8
33960/pWLfdxC	11.3	11.2	12.3	16.7
33960/pWLEC	13.6	14.6	15.4	18.0

Both phaE and phaC genes are necessary for an active PHA synthase in Haloarcula.

To further clarify the functions of the phaEC_Hm genes, the PHA synthase activities in the phaEC_Hm-overexpressing H. hispanica strains were assayed. Compared with that in the control strain H. hispanica ATCC 33960/pWL102 (∼1.0 U/mg protein), it was revealed that the PHA synthase activities of the crude extracts from H. hispanica recombinants harboring phaE_Hm (33960/pWLE; ∼2.0 U/mg protein) or phaC_Hm (33960/pWLfdxC; ∼1.6 U/mg protein) were clearly increased. Again, coexpression of phaEC_Hm genes in H. hispanica (33960/pWLEC) resulted in the highest PHA synthase activity (∼2.6 U/mg protein). These results suggested that the phaEC genes might encode the same subunits of PHA synthase in Haloarcula as those observed in bacteria. To confirm this, we knocked out the phaEC_Hh genes from H. hispanica ATCC 33960 (see Materials and Methods) and thus generated a phaEC-deleted strain named H. hispanica PHB-1. As expected, H. hispanica PHB-1 completely lost the PHA synthase activity and the capability for PHB synthesis when cultivated in MG medium. This result supported the argument that the phaEC genes might encode the PHA synthase in the Haloarcula species. To further evaluate the function of PhaE and PhaC in PHB synthesis, the plasmids pWLE, pWLfdxC, and pWLEC were also transformed into H. hispanica PHB-1. It was revealed that neither pWLE nor pWLfdxC, but only pWLEC, could restore the PHA synthase activity (data not shown). Consistently, only coexpression of phaEC_Hm genes in H. hispanica PHB-1 restored the ability to accumulate PHB (Fig. 6C). Expression of either phaE_Hm or phaC_Hm in H. hispanica PHB-1 did not result in PHB accumulation (Fig. 6A and B). These results strongly suggested that the phaEC genes encode two subunits, PhaE and PhaC, of a novel class III PHA synthase in the Haloarcula species.

FIG. 6.

GC analysis of PHB accumulation in H. hispanica PHB-1 recombinants. The 3HB peaks are the actual peaks of 3-hydroxybutyrate methylester, the methanolysis product of PHB (see Materials and Methods). (A) H. hispanica PHB-1 harboring pWLE; (B) H. hispanica PHB-1 harboring pWLfdxC; (C) H. hispanica PHB-1 harboring pWLEC; (D) PHB standard (Sigma). The peak at 4.85 min represents the methylester product of an internal standard (1 ng of benzoic acid). The H. hispanica PHB-1 recombinants were cultivated in MG medium at 37°C for 96 h.

DISCUSSION

In the present study, we report for the first time the identification and molecular characterization of two genes, phaE_Hm and phaC_Hm, that are essential for both PHB accumulation and PHA synthase activity in H. marismortui ATCC 43049. We have determined their genetic organization, studied their regulation at the transcriptional and translational level, and proved by synthase activity assay and PHB accumulation analysis in a heterologous system as well as a gene knockout/complementation system that the phaEC genes might encode two subunits of an active PHA synthase in Haloarcula.

It is noteworthy that the PHA synthase in H. marismortui was likely a homolog of the bacterial class III synthase, as both PhaE_Hm and PhaC_Hm contained conserved motifs typically found in class III PHA synthases from Bacteria (Fig. 3). Similarly, the phaE_Hm and phaC_Hm genes were controlled by a single promoter (Fig. 4), which also suggested that PhaE_Hm and PhaC_Hm might function together in H. marismortui. However, the protein sizes of PhaE_Hm (20.64 kDa) and PhaC_Hm (53.12kDa) were distinct from typical class III PHA synthases of bacteria. e.g., the PhaE_Av and PhaC_Av of A. vinosum were about 40.53 kDa and 39.73 kDa, respectively (22). Moreover, the stretch rich in hydrophobic and amphiphilic amino acids for granule association in most bacterial PhaE subunits (20) was absent from the C terminal of PhaE_Hm. These results indicated that the PHA synthase of H. marismortui was a novel form and may explain why Western blotting could only detect PhaC_Hm but not PhaE_Hm associated with PHB granules (Fig. 5). It is possible that PhaE_Hm combined with PhaC_Hm with a weak strength to form an active complex enzyme, and the conditions of purifying the PHB granules might have released PhaE_Hm. Besides PhaE and PhaC from the genus Haloarcula, the PhaC_Hw subunit from another genus of haloarchaea, H. walsbyi DSM 16790, also showed high homology with the class III PhaC (Fig. 3), thereby indicating that the class III PHA synthases might be widely spread from Bacteria to Archaea.

H. marismortui ATCC 43049 accumulated PHB when it was cultured in minimal medium (pH 7.2) with excess glucose (Fig. 1). But, surprisingly, we revealed that the phaEC_Hm genes were constitutively expressed, no matter whether the cells were cultured under nutrient-rich (AS-168 medium) or nutrient-limited (MG medium) conditions (Fig. 4 and 5). This was different from results with the haloarchaeal strain 56, whose PHA synthase activity presented only under PHB-accumulating conditions (11), but resembled many bacteria, including Cupriavidus necator, A. vinosum, Synechococcus sp., and Paracoccus denitrificans, whose PHA synthases are also constitutively expressed (16, 21, 35). This constitutive production of the PHA synthases may be an important mechanism for microorganisms to synthesize PHAs as soon as they sense the imbalance of nutrients.

In addition, we overexpressed either or both of the phaE_Hm and phaC_Hm genes in H. hispanica ATCC 33960 harboring a highly homologous phaEC_Hh operon, and we found only coexpression of phaE_Hm and phaC_Hm genes resulted in the highest PHB content and PHB yield (Table 3) in these transformants. We also introduced phaE_Hm and/or phaC_Hm genes into a phaEC-deleted mutant strain, H. hispanica PHB-1, and revealed that only coexpression of phaE_Hm and phaC_Hm could restore H. hispanica PHB-1 with the ability to synthesize PHB as well as PHA synthase activity. These results clearly confirmed the conclusion that PhaE and PhaC should be the active components of the PHA synthase in Haloarcula. However, introduction of phaEC_Hm genes into the PHA synthase-negative bacterium Cupriavidus necator PHB-4 (32) could not restore its ability to accumulate PHB, although the expression of PhaE_Hm and PhaC_Hm was confirmed by Western blotting (data not shown), suggesting that the PHA synthase of H. marismortui ATCC 43049 might not be able to fold into its native conformation in the low-salt cytoplasm of the bacterium. It would be interesting to investigate the structure and biochemical features of these novel haloarchaeal PHA synthases in the future.