Phasin Proteins Activate Aeromonas caviae Polyhydroxyalkanoate (PHA) Synthase but Not Ralstonia eutropha PHA Synthase

Abstract

In this study, we performed in vitro and in vivo activity assays of polyhydroxyalkanoate (PHA) synthases (PhaCs) in the presence of phasin proteins (PhaPs), which revealed that PhaPs are activators of PhaC derived from Aeromonas caviae (PhaC_Ac). In in vitro assays, among the three PhaCs tested, PhaC_Ac was significantly activated when PhaPs were added at the beginning of polymerization (prepolymerization PhaC_Ac), whereas the prepolymerization PhaC_Re (derived from Ralstonia eutropha) and PhaC_Da (Delftia acidovorans) showed reduced activity with PhaPs. The PhaP-activated PhaC_Ac showed a slight shift of substrate preference toward 3-hydroxyhexanoyl-CoA (C₆). PhaP_Ac also activated PhaC_Ac when it was added during polymerization (polymer-elongating PhaC_Ac), while this effect was not observed for PhaC_Re. In an in vivo assay using Escherichia coli TOP10 as the host strain, the effect of PhaP_Ac expression on PHA synthesis by PhaC_Ac or PhaC_Re was examined. As PhaP_Ac expression increased, PHA production was increased by up to 2.3-fold in the PhaC_Ac-expressing strain, whereas it was slightly increased in the PhaC_Re-expressing strain. Taken together, this study provides evidence that PhaPs function as activators for PhaC_Ac both in vitro and in vivo but do not activate PhaC_Re. This activating effect may be attributed to the new role of PhaPs in the polymerization reaction by PhaC_Ac.

INTRODUCTION

Polyhydroxyalkanoates (PHAs) are aliphatic polyesters that are synthesized and accumulated by a wide range of microorganisms as their carbon- and energy-storage materials (1, 2). PHAs have attracted attention as biodegradable thermoplastics for industrial applications. A key enzyme in PHA biosynthesis is PHA synthase (PhaC), which is classified into four groups (classes I to IV) based on substrate specificity and subunit composition (see the review by Rehm [1] for a detailed classification of PHA synthases).

Class I PhaCs are homomeric enzymes comprised of PhaC subunits with molecular mass of 60 to 70 kDa; these enzymes prefer to polymerize monomers with a short acyl chain length (C₃ to C₅). As a typical class I enzyme, Ralstonia eutropha PHA synthase (PhaC_Re) has been well studied and characterized. Delftia acidovorans synthase (PhaC_Da) is also a class I synthase but has a unique insertion sequence of 40 amino acids located at the C terminus of the active center cysteine at position 322 (3). PhaC derived from Aeromonas caviae (PhaC_Ac) is not a typical class I enzyme because it is able to polymerize not only the C₄ and C₅ substrates but also C₆ substrates. PhaC_Ac shows a relatively low amino acid sequence identity to PhaC_Re (37.4%) and PhaC_Da (31.1%), whereas PhaC_Re and PhaC_Da are relatively similar (50.7%).

A number of enzymes and nonenzymatic proteins are involved in PHA biosynthesis. Phasin proteins (PhaPs), which are PHA granule-associated proteins localized on the surfaces of the PHA granules in bacterial cells (4, 5), are well known PHA synthesis-related proteins. The primary function of PhaPs is control of the surface properties of PHA granules. PhaPs bind strongly to the hydrophobic surfaces of growing PHA granules to block the binding of other proteins. Accordingly, PhaP expression levels in bacterial cells affect the hydrophobicity of the surface of PHA granules, thereby influencing granule development via hydrophobic aggregation of small granules (4, 6).

Most phasins, including R. eutropha phasin (PhaP1_Re), have a molecular mass of approximately 20 kDa (5). The A. caviae phasin (PhaP_Ac) is a small protein having a molecular mass of 13 kDa and low identity to PhaP1_Re (13.1%) (7). Several interesting observations suggested a second function of PhaP_Ac. These observations include PhaP_Ac expression leading to enhanced accumulation of PHA and/or altering the copolymer composition, likely by influencing the activity and substrate specificity of PhaC_Ac (8, 9). The same phenomenon was observed in recombinant bacteria between the PhaC and PhaP derived from Aeromonas hydrophila (10). However, these in vivo studies have not focused on the role of PhaP at the molecular level.

The aim of our study was to investigate the role of PhaP in the modulation of PhaC activity. For this purpose, we used purified proteins to assess the in vitro activity of PhaCs derived from A. caviae, R. eutropha, and D. acidovorans in the presence of PhaPs derived from A. caviae and R. eutropha. In vivo PHA production was also conducted using PhaCs from A. caviae and R. eutropha with or without the expression of PhaP_Ac. Based on the in vivo and in vitro results, we propose a new physiological role of PhaPs in the polymerization reaction catalyzed by PhaC_Ac.

MATERIALS AND METHODS

Plasmid construction.

Plasmids to express His-tagged PhaC_Ac, PhaP_Ac, and PhaP1_Re were constructed based on the pColdI vector (TaKaRa Bio, Inc., Otsu, Japan) to obtain purified proteins. For construction of pColdI::phaC_Ac and pColdI::phaP_Ac, phaC_Ac and phaP_Ac genes were obtained by PCR with pBBR1phaPCJ_AcAB_Re (11) as the template. The following PCR primers were used: forward for phaC_Ac, 5′-GGGTGAAGGAGAGCATATGAGCCAACCATCTTATGG-3′, reverse for phaC_Ac, 5′-GACCACGGATCCTGCGCTCA-3′, forward for phaP_Ac 5′-TGGAGACCGCATATGAATATGGACGTGATC-3′, and reverse for phaP_Ac, 5′-CAGGGATCCTCAGGCCTTGCCCGTGCTTTT-3′ (underlined sequences indicate the NdeI and BamHI sites, respectively). These gene fragments were digested by NdeI and BamHI and were introduced into the same sites of the pColdI vector. For pColdI::phaP1_Re, genomic DNA of R. eutropha H16 was used as the template. The following PCR primers were used: forward, 5′-ACTGGAGACCACATATGATCCTCACCCCGG-3′, and reverse, 5′-GGATCCTCAGGCAGCCGTCGTCTTCTTTGC-3′ (underlined sequences indicate the NdeI and BamHI sites, respectively). The amplified DNA fragment carrying the phaP1_Re gene was digested by NdeI and BamHI and was introduced into the same sites of the pColdI vector. Three plasmids, named pBAD::TEE+phaP_Ac, pBBR1::phaC_AcAB_Re, and pBBR1::phaCAB_Re, were newly constructed for in vivo PHA production experiments. Maps of the plasmids used for in vivo experiments are shown in Fig. 1. In the construction of pBAD::TEE+phaP_Ac, the phaP_Ac gene fragment, which includes the Shine-Dalgarno sequence, translation-enhancing element (TEE) sequence, and His tag sequence of pColdI::phaP_Ac, was amplified by PCR using pColdI::phaP_Ac as the template. The following PCR primers were used: forward, 5′-TTAGAGCTCAAGAGGTAATACACCATGAAT-3′, and reverse, 5′-TTCGGTACCTCAGGCCTTGCCCGTGCTTTT-3′ (underlined sequences indicate the SacI and KpnI sites, respectively). For pBBR1::phaC_AcAB_Re and pBBR1::phaCAB_Re, the BamHI-digested fragments from pGEM′-phbCAB_Re (12) and pGEM-phaC_AcAB (13), respectively, were ligated with BamHI-digested pBBR1-MCS2 (14). These PHA biosynthesis genes were located downstream from the promoter of the PHA biosynthesis operon (phaCAB_Re) from R. eutropha, and they were directionally opposite from the lac promoter present in pBBR1-MCS2.

FIG 1.

Map of plasmids used for in vivo experiments. (a) pBBR1-MCS2-derived plasmid for PHA biosynthesis. (b) pBAD33-derived plasmid for PhaP_Ac expression.

Preparation of PhaCs and PhaPs.

PhaC_Ac with the His tag removed was prepared by using cell-free protein enzymes as reported previously (15). PhaC_Ac was eluted with 20 mM Tris-HCl buffer (pH 7.5) containing 150 mM NaCl. His-tagged PhaC_Re and PhaC_Da were expressed in the recombinant Escherichia coli BL21(DE3) strain (Novagen, Madison, WI) with the pET-15b::phaC_Re and pET-15b::phaC_Da vectors, respectively (16, 17). PhaC_Re and PhaC_Da were purified from the cells using a HisTrap HP column (GE Healthcare), followed by desalting using a HiTrap desalting column with desalting buffer [20 mM Tris-HCl, 150 mM NaCl, 5% (vol) glycerol, and 0.05% (wt) 6-O-(N-heptylcarbamoyl)methyl α-d-glucopyranoside (Hecameg), pH 7.5]. All processes were conducted at 4°C. Aliquots of purified PhaCs were frozen in liquid nitrogen and stored at −80°C.

His-tagged PhaC_Ac, PhaP_Ac, and PhaP1_Re were expressed in the recombinant E. coli BL21(DE3) strain with pColdI::phaC_Ac, pColdI::phaP_Ac, and pColdI::phaP1_Re vectors, respectively. All strains were grown in Luria-Bertani (LB) broth (1% Bacto-tryptone, 0.5% yeast extract, and 1% NaCl, pH 7.0) with 100 μg/ml ampicillin. These expression strains were grown at 37°C to an optical density at 600 nm (OD₆₀₀) of ≈0.6 with shaking. The cells were incubated at 16°C for 30 min without shaking, and 0.1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) was added. After the addition of IPTG, the cells were cultivated at 16°C for 24 h with shaking. Proteins were purified using a HisTrap HP column (GE Healthcare). Finally, purified PhaC_Ac was desalted using desalting buffer, as described for PhaC_Re and PhaC_Da; PhaPs were desalted using ultrapure water. These purified and desalted proteins were frozen in liquid nitrogen and stored at −80°C. All proteins were purified to near homogeneity (see Fig. S1 in the supplemental material). Protein concentrations were determined using a Quant-iT protein assay kit (Invitrogen, Carlsbad, CA).

Monomer preparation.

To assay PhaC activity, (R)-3-hydroxybutyryl-coenzyme A (R-3HB-CoA) and (R)-3-hydroxyhexanoyl-CoA (R-3HHx-CoA) were prepared as substrates. R-3HB-CoA was chemically synthesized and purified as previously described (18). R-3HHx-CoA was synthesized from hexenoyl-CoA using A. caviae (R)-specific enoyl-CoA hydratase as previously reported (19) and purified using high-performance liquid chromatography (HPLC) in the same manner as for R-3HB-CoA. Substrate concentrations were quantified relative to an HPLC standard curve of racemic 3HB-CoA (Sigma; St. Louis, MO).

Activity assay of PhaCs in the presence of PhaPs.

For the PhaC activity assay, a decrease in the absorbance at 236 nm, which occurs due to the cleavage of the thioester bond in R-3HB-CoA (ε₂₃₆ = 4,500 [M⁻¹ · cm⁻¹]), was measured (20). The assay was carried out for the three PhaCs in the presence or absence of the PhaPs at 30°C in 50 mM sodium phosphate buffer (pH 7.0) with R-3HB-CoA or R-3HHx-CoA. Detailed assay conditions are described below. Specific activities were determined using the maximum velocities of each reaction.

PHA biosynthesis in vivo.

For poly[(R)-3-hydroxybutylate] [P(3HB)] synthesis, E. coli TOP10 cells (Invitrogen, Carlsbad, CA) harboring a pBBR1-MCS2-derived plasmid (pBBR1::phaC_AcAB_Re or pBBR1::phaCAB_Re) and pBAD::TEE+phaP_Ac (or the pBAD33 empty vector as a control) were cultivated as described below. These strains were inoculated into LB medium (containing 50 μg/ml kanamycin and 30 μg/ml chloramphenicol) in test tubes and cultured at 37°C for 15 h. These cultures (1 ml) were inoculated into 500-ml shake flasks containing 100 ml LB medium, 50 μg/ml kanamycin, 30 μg/ml chloramphenicol, 20 g/liter d-glucose, and 0 to 1% (wt/vol) l-arabinose. These flasks were cultivated at 37°C for 72 h using a reciprocal shaker (130 strokes/min). After cultivation, cells were centrifuged, washed with pure water, and lyophilized.

Analyses of PHA produced in vivo.

The PHA content in the cells was determined by gas chromatography (GC) after methanolysis in the presence of 15% (vol/vol) sulfuric acid, as described previously (21). Synthesized PHA was extracted from cells by stirring with chloroform for 72 h at room temperature, followed by filtration of cell debris. The extracted PHA solutions were precipitated into methanol.

The molecular weights and their distribution were determined by gel permeation chromatography (GPC) at 40°C using a Shimadzu 10A GPC system equipped with a 10A refractive index detector with two Shodex K-806 M joint columns. Chloroform was used as the eluent, with a flow rate of 0.8 ml/min; the sample concentrations were 1.0 mg/ml. Low polydispersity polystyrenes were used as the molecular-weight standard. All PHA samples for molecular-weight measurements were obtained from cultivations with 1.0% (wt/vol) l-arabinose.

Western blotting of PhaCs.

The amounts of PhaC_Ac and PhaC_Re in cells were analyzed by Western blotting as described previously (17). Briefly, E. coli cells were cultivated in the LB-plus-glucose medium described above (containing 1.0% [wt/vol] l-arabinose), and the cells were harvested after 6 h of cultivation and disrupted by sonication. The disrupted cells were centrifuged at low speed (1,500 × g for 5 min at 4°C) to separate the soluble cell extract containing PHA granules and an insoluble precipitate. The protein concentrations in the soluble/insoluble fractions were determined using the Quant-iT protein assay kit (Invitrogen). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed using 8 μg of protein in each soluble/insoluble sample. The detection of PhaC_Ac and PhaC_Re was carried out with specific rabbit antisera to PhaC_Ac (22) and PhaC_Re (23), respectively. Protein bands were visualized using goat anti-rabbit antibodies conjugated with horseradish peroxidase (HRP) (Santa Cruz Biotechnology, CA, USA) as a secondary antibody and an ECL (enhanced chemiluminescence) plus Western blotting detection reagent (GE Healthcare).

RESULTS

Activity assay of PhaCs in the presence of PhaPs in vitro.

In vitro polymerization activity assays using purified PhaCs and PhaPs were conducted. The results of the activity assay are shown in Table 1. In the PhaP-free assay, PhaC_Ac without a His tag showed the lowest activity (0.76 U/mg PhaC) toward R-3HB-CoA among the PhaCs examined, while PhaC_Re showed the highest activity (3.51 U/mg PhaC). Interestingly, the addition of 1 μM PhaP_Ac into the reaction mixture exhibited two different effects on PhaC activity; the activity of PhaC_Ac increased to 2.31 U/mg PhaC in the presence of PhaP_Ac, which was 3.0-fold higher than that in the PhaP-free assay. In contrast, the addition of 1 μM PhaP_Ac decreased the activities of PhaC_Re and PhaC_Da by approximately 10-fold; the calculated activities of PhaC_Re and PhaC_Da were 0.36 U/mg PhaC and 0.15 U/mg PhaC, respectively, in the presence of PhaP_Ac.

TABLE 1.

PhaC activity assay toward R-3HB-CoA in the presence of PhaP^a

Assay conditions	Sp act (mean U/mg PhaC ± SE) (fold activation by PhaP) of:
	PhaCAc		PhaCRe	PhaCDa
	Without His tag	With His tag
PhaP free	0.76 ± 0.08 (1.0)	0.72 ± 0.03 (1.0)	3.51 ± 0.21 (1.0)	1.40 ± 0.14 (1.0)
+PhaPAc	2.31 ± 0.07 (3.0)	2.24 ± 0.11 (3.1)	0.36 ± 0.07 (0.1)	0.15 ± 0.03 (0.1)
+PhaP1Re	1.82 ± 0.10 (2.4)	1.76 ± 0.03 (2.4)	0.39 ± 0.02 (0.1)	0.64 ± 0.06 (0.5)

^aAssay conditions: PhaCs, 350 nM (PhaC_Ac with/without His tag) or 200 nM (PhaC_Re and PhaC_Da with a His tag); R-3HB-CoA, 100 μM; PhaP, 0 or 1 μM. n = 3 replicates.

Furthermore, a similar effect was observed by adding another phasin, PhaP1_Re. In the presence of 1 μM PhaP1_Re, PhaC_Ac activity increased to 1.82 U/mg PhaC, whereas PhaC_Re and PhaC_Da activities decreased to 0.39 U/mg PhaC and 0.64 U/mg PhaC, respectively. These results suggest that PhaPs function as modulators of PhaC activity but that the modulatory action (activating or inhibiting) is PhaC dependent. Namely, PhaPs enhance the activity of PhaC_Ac but inhibit the activity of PhaC_Re and PhaC_Da. This trend was also observed in the parallel assay using His-tagged PhaC_Ac purified from E. coli BL21(DE3) cells harboring pColdI::phaC_Ac (Table 1).

Effect of PhaP concentration on PhaC_Ac activation in vitro.

We assayed the activity of PhaC_Ac relative to its activity under the PhaP-free condition by varying the concentrations of PhaPs added into the reaction mixture (Fig. 2; reaction progress curves are shown in Fig. 3). PhaC_Ac activity increased 3.0-fold with increasing PhaP_Ac concentration up to 1 μM; a further increase in the activity of PhaC_Ac was not observed with the addition of 5 μM PhaP_Ac. These data suggest that the activating effect of PhaP_Ac reached a plateau around a concentration of 1 μM, where the molar ratio of PhaP_Ac to PhaC_Ac was 2.9 (mol/mol).

FIG 2.

Relative activities of PhaC_Ac and PhaC_Re for the R-3HB-CoA substrate at different concentrations of PhaP_Ac and PhaP1_Re. Relative activities were determined using the change in absorbance at 236 nm at maximum velocity. Reaction mixtures containing PhaC (350 nM PhaC_Ac without a His tag or 200 nM PhaC_Re), 100 μM R-3HB-CoA, and 0 to 5 μM PhaP (PhaP_Ac or PhaR1_Re) were used. Results are expressed as the means ± standard errors (n = 3), except for results with 5 μM PhaP (n = 2).

FIG 3.

Effects of PhaP_Ac on the PhaC_Ac reaction rate for the R-3HB-CoA substrate. The assay was carried out by monitoring the changes in absorbance at 236 nm of reaction mixtures containing 100 μM R-3HB-CoA, 350 nM PhaC_Ac without a His tag, and the indicated concentrations of PhaP_Ac.

The effect of another phasin, PhaP1_Re, on the activity of PhaC_Ac was also investigated (Fig. 2). PhaP1_Re increased PhaC_Ac activity up to 2.7-fold compared to its activity in the absence of PhaP; however, the activating effect was slightly lower than that of PhaP_Ac. The enhancing effect reached a plateau at around 1 μM PhaP1_Re (the molar ratio of PhaP1_Re to PhaC_Ac was 2.9 [mol/mol]), which is identical to that of PhaP_Ac, regardless of the molecular mass of the PhaP. At the initial stage of polymerization, the maximum polymerization activity may be achieved at a PhaP/PhaC_Ac molar ratio of 3:1.

Reactivity of activated PhaC_Ac toward 3HHx in vitro.

PhaC_Ac that was activated by 2.5 μM PhaP_Ac was spectrophotometrically assayed using R-3HHx-CoA (C₆ substrate) and R-3HB-CoA (C₄ substrate). Table 2 shows the specific activities of PhaC_Ac with a high concentration (150 μM) of C₄ and C₆ substrates, which clearly show a weak activity toward the C₆ substrate. The activities for the C₄ and C₆ substrates were increased by the addition of 2.5 μM PhaP_Ac. The specific activities of the C₆ substrate in the presence and absence of PhaP_Ac were 0.69 U/mg PhaC and 0.16 U/mg PhaC, respectively. This activation fold increase was calculated to be 4.3, which was higher than that of the C₄ substrate (2.5-fold), and it resulted in a higher C₆/C₄ activity ratio. These data indicate that PhaP_Ac increased the specificity of PhaC_Ac against the C₆ substrate compared to its activity against the C₄ substrate.

TABLE 2.

PhaC_Ac activity toward C₄ and C₆ substrates in vitro^a

Assay conditions	Sp act (mean U/mg PhaCAc ± SE) (fold activation by PhaP) with:		C6/C4 activity ratio
	R-3HB-CoA (C4)	R-3HHx-CoA (C6)
PhaCAc	1.27 ± 0.03 (1.0)	0.16 ± 0.01 (1.0)	0.13
PhaCAc + PhaPAc	3.15 ± 0.13 (2.5)	0.69 ± 0.03 (4.3)	0.22

^aAssay conditions: PhaC_Ac without a His tag, 700 nM; substrate (R-3HB-CoA or R-3HHx-CoA), 150 μM; PhaP_Ac, 0 or 2.5 μM. n = 3 replicates.

Effect of PhaP concentration on PhaC_Re inhibition in vitro.

The activities of PhaC_Re and PhaC_Da were inhibited by 1 μM PhaP (Table 1), which was in contrast to the activation observed for PhaC_Ac. To investigate the effect of PhaP concentration, activity assays of PhaC_Re with various PhaP concentrations were conducted (Fig. 2; the reaction progress curves are shown in Fig. 4). The inhibitory effect on PhaC_Re activity was dependent on the concentration of the PhaPs, and it reached a plateau at around 0.5 μM PhaP. PhaP_Ac showed a slightly stronger inhibitory effect than PhaP1_Re (Table 1 and Fig. 2).

FIG 4.

Effects of PhaP_Ac on the PhaC_Re reaction rate for the R-3HB-CoA substrate. The assay was carried out by monitoring the changes in absorbance at 236 nm of reaction mixtures containing 100 μM R-3HB-CoA, 200 nM PhaC_Re, and the indicated concentrations of PhaP_Ac.

Effect of PhaP on the activity of polymer-elongating PhaC in vitro.

We conducted activity assays of polymer-elongating PhaCs in the presence or absence of PhaP_Ac. In this assay, the polymerization reaction by PhaC_Ac (350 nM) or PhaC_Re (200 nM) was started with 50 μM R-3HB-CoA. After 300 s of incubation with almost no change in absorbance at 236 nm, 1 μM PhaP_Ac was added to the reaction mixture, followed by an additional 50 μM R-3HB-CoA. The progress curves of this assay using PhaC_Ac and PhaC_Re are shown in Fig. 5. An activation effect of PhaP_Ac on the polymer-elongating PhaC_Ac was observed, similar to the prepolymerization PhaC_Ac (Table 1 and Fig. 2). On the other hand, polymer-elongating PhaC_Re was not affected by the presence of PhaP_Ac (Fig. 5b), unlike the prepolymerization PhaC_Re, which was inhibited by PhaP_Ac (Fig. 2). The relative activities (U/mg PhaC) of polymer-elongating PhaCs were as follows: PhaC_Ac alone, 0.66 ± 0.03 (mean ± standard deviation); PhaC_Ac plus PhaP_Ac, 2.07 ± 0.11; PhaC_Re alone, 1.74 ± 0.12; PhaC_Re plus PhaP_Ac, 1.65 ± 0.12.

FIG 5.

Reaction progress curves obtained with 350 nM PhaC_Ac with a His tag (a) and 200 nM PhaC_Re (b) during two-step addition of R-3HB-CoA. Reactions were initiated with 50 μM R-3HB-CoA, followed by the addition of PhaP_Ac (1 μM in final concentration) or buffer (control) and 50 μM R-3HB-CoA with a 300-s incubation.

Effect of PhaP on PHA production in vivo.

To evaluate the effect of PhaP on PHA production in vivo, E. coli TOP10 cells harboring P(3HB) biosynthesis genes (phaC_Ac/phaC_Re, phaA_Re, and phaB_Re) under pha_Re promoter control and the phaP_Ac gene under l-arabinose-inducible promoter control were cultured in LB medium containing 20 g/liter glucose (Fig. 6). The P(3HB) content was increased at high concentrations of l-arabinose (0.2% or 1.0% [wt/vol]) in both PhaC_Ac- and PhaC_Re-expressing strains. The maximum P(3HB) content was 2.3-fold (PhaC_Ac) or 1.2-fold (PhaC_Re) higher than the levels without PhaP_Ac expression. The molecular weights of P(3HB) extracted from 1.0% l-arabinose culture are listed in Table 3. These molecular weights were reduced in both strains by the expression of PhaP_Ac. In particular, the PhaC_Re-expressing strain exhibited a remarkable reduction in molecular weight. Western blotting was performed to estimate the amounts of PhaCs in the soluble fractions obtained by low-speed centrifugation (Fig. 7). Based on the band intensity, the amount of soluble PhaC_Ac was decreased to one-half by the expression of PhaP_Ac (Fig. 7a), while the amount of PhaC_Re in the soluble fraction was increased approximately 3-fold by PhaP_Ac expression (Fig. 7b).

FIG 6.

Effect of PhaP_Ac expression on P(3HB) accumulation in the PhaC_Ac-expressing strain (pBBR1::phaC_AcAB_Re) (a) and the PhaC_Re-expressing strain (pBBR1::phaCAB_Re) (b). A higher l-arabinose concentration induced a stronger expression of PhaP_Ac in the araBAD system. Empty plasmid pBAD33 was retained in each control strain instead of using pBAD::TEE+phaP_Ac. Residual cell mass was calculated as the dry cell weight minus P(3HB). Error bars indicate the standard deviations (n = 3).

TABLE 3.

P(3HB) synthesis in recombinant E. coli TOP10 cells cultured in the presence of l-arabinose^a

Expressed protein(s)	Dry cell wt (g/liter)	P(3HB) content (wt%)	Mn (×105)	Mw (×105)	Mw/Mn
PhaCAc	1.2 ± 0.1	17 ± 2	3.5 ± 0.6	12.0 ± 0.9	3.4
PhaCAc + PhaPAc	2.1 ± 0.4	39 ± 4	2.4 ± 0.6	7.4 ± 0.5	3.1
PhaCRe	1.9 ± 0.1	35 ± 2	24.9 ± 3.5	39.9 ± 6.3	1.6
PhaCRe + PhaPAc	2.1 ± 0.1	43 ± 1	9.3 ± 1.1	24.4 ± 1.4	2.6

^aStrains were the same as in the experiments whose results are shown in Fig. 6. The l-arabinose concentration was 1.0% (wt/vol). Results are expressed as the means ± standard errors (n = 3).

FIG 7.

Western blots of PhaCs in 6-h-cultured PhaC_Ac-expressing strains (a) and PhaC_Re-expressing strains (b) with and without PhaP expression. The l-arabinose concentration was 1.0% (wt/vol). Lanes 1, 3, 5, 7, 9, and 11 are PhaP_Ac-free strains. Lanes 2, 4, 6, 8, 10, and 12 are PhaP_Ac-expressing strains. The soluble cell extract, containing PHA granules, and the insoluble precipitate were separated by low-speed centrifugation (1,500 × g, 5 min, 4°C). The amount of protein in each lane was 8 μg.

DISCUSSION

Previous studies have indicated that PhaPs have an enhancing effect on PHA biosynthesis in vivo (9, 10, 24, 25). To further investigate this effect, we performed in vitro polymerization activity assays in this study. Our results demonstrated that both PhaP_Ac and PhaP1_Re increased PhaC_Ac activity, providing convincing evidence of PhaP-mediated PhaC_Ac activation at the molecular level. In contrast, PhaC_Re and PhaC_Da showed reduced activities in the presence of PhaPs (Table 1). Some detergents, such as Triton X-100 and Hecameg, have been shown to activate PhaC_Re and PhaC_Da in vitro (17, 18). We tested the ability of these detergents to activate PhaC_Ac, but an activating effect was not observed (data not shown). Although these detergents and PhaPs are both amphiphilic molecules, the behavior of PhaCs was very different depending on the additive used. Therefore, PhaPs and the detergents appear to have different activation mechanisms for PhaCs. Most recently, PhaM, a DNA- and PHA-binding protein, has been shown to have an activating effect for PhaC_Re (26). PhaPs and PhaM have the same ability to bind PHA granules; however, their effects on PhaC_Re activity were opposite. The underlying mechanism for this difference between PhaPs and PhaM would be an interesting subject for study.

With respect to P(3HB-co-3HHx) production in the PhaC_Ac-expressing strain, several studies have reported that PhaP_Ac affects not only PHA accumulation levels but also the 3HHx fraction in the polymer (8, 9). We hypothesized that the enhanced 3HHx fraction was also the result of the PhaP-activating effect of PhaC_Ac. To test this hypothesis, the effect of PhaP_Ac on the substrate specificity of PhaC_Ac was investigated using R-3HHx-CoA (C₆ substrate) and R-3HB-CoA (C₄ substrate). The PhaC_Ac activity toward R-3HHx-CoA was increased 4.3-fold by the presence of PhaP_Ac, which was a significantly higher fold increase in activation than was observed for the C₄ substrate (2.5-fold), suggesting that the substrate preference of the activated PhaC_Ac is slightly shifted toward the C₆ substrate. This shift may explain the increased 3HHx incorporation previously observed in vivo (8, 9).

Regarding the activity assays of PhaC_Re in the presence of PhaPs, the in vitro results were not consistent with the in vivo result (Fig. 2 and 6b). In the presence of PhaPs, PhaC_Re showed reduced activity in vitro, but the activity was not reduced in vivo. Cho and coworkers (27) reported a similar inhibitory effect of PhaP1_Re on PhaC_Re activity in vitro. According to their report (27), PhaP1_Re reduced the in vitro activity of PhaC_Re mainly by increasing the lag phase at the start of polymerization. However, our observations indicated that the maximum velocity in the activity assay was mainly repressed by the presence of PhaPs (Fig. 4). The reason for this difference might be attributed to the difference in PhaP1_Re concentrations; namely, the PhaP concentration in our study (5 μM) was higher than the concentration used in their study (0.14 μM).

In addition, Cho and coworkers (27) reported that the PhaC_Re-PHA-PhaP1_Re complex showed higher polymerization activity than the prepolymerization PhaC_Re even though the PhaC_Re-PHA-PhaP1_Re complex contained a high molar ratio of PhaP1_Re (PhaP1_Re-PhaC_Re-PHA chain = 9:3:1). This observation has led to the hypothesis that the PHA chain prevents the inhibitory effect of PhaPs on PhaC_Re polymerization activity. To test this hypothesis, we conducted an activity assay using polymer-elongating PhaCs. The polymer-elongating PhaC_Re prepared by the two-step addition of substrate in vitro showed almost the same polymerization activity regardless of PhaP addition (Fig. 5b), suggesting that the PHA chain plays an important role in preventing the reduction of PhaC_Re activity by PhaPs. The inhibitory effect by PhaP only occurred in the initial stage of the polymerization reaction; therefore, it is not likely to be observed in in vivo studies.

As shown in Fig. 6, the PhaP_Ac expression level was controlled by the l-arabinose concentration. The P(3HB) content was increased in both PhaC_Ac- and PhaC_Re-expressing strains when the l-arabinose concentrations were over 0.2% (wt/vol). The highest P(3HB) accumulations were observed at 1.0% (wt/vol) l-arabinose, leading to 2.3- and 1.2-fold increases in the P(3HB) content for the PhaC_Ac- and PhaC_Re-expressing strains, respectively, compared with the expression levels in the corresponding PhaP_Ac-free strains. These data imply that a higher expression level of PhaP_Ac results in a greater positive effect on PHA production in vivo. From the Western blot analysis of 6-h-cultured cells (Fig. 7), the amount of soluble PhaC_Ac was slightly decreased by the expression of PhaP_Ac, while the amount of soluble PhaC_Re was markedly increased. The difference in the levels of soluble PhaC_Ac would result from differences in the cell growth phase, that is, the PhaP_Ac-free strain was still actively growing at 6 h of cultivation (polymer content, 9% [wt]), while the PhaP_Ac-expressing strain was already beginning to accumulate P(3HB) (22% [wt]) due to the activated PhaC_Ac. Thus, the protein expression pattern in these PhaC_Ac-expressing strains might be slightly different. On the other hand, the difference in soluble PhaC_Re is attributed to the increased solubility of PhaC_Re by the presence of PhaP_Ac. In general, as the amount of soluble PhaC_Re increases in cells, the PHA chain number increases, whereas the molecular weight decreases (23). The P(3HB) molecular weight data (Table 3) are consistent with the fact that the amount of soluble PhaC_Re increased. These observations may suggest that PhaP_Ac plays a chaperone-like role in PhaC_Re folding, which is a previously unrecognized function of PhaPs. Further studies are needed to confirm the role of PhaPs in protein folding.

PhaPs have been reported to form trimers or tetramers as shown by X-ray analysis (28, 29). On the other hand, PhaCs were observed in oligomeric form at the initial stage of PHA polymerization by atomic force microscope imaging (19, 30), but the PhaC dimer is thought to be a minimal functional unit (1, 15). Based on the observations in this study and previous works (19, 27–30), we propose a role for PhaPs in the initial stage of PHA polymerization by PhaC_Ac, as shown in Fig. 8. In this model, we assume that PHA synthase has a monomer uptake site and a polymer release site, which are distantly located. In the PhaP-free model, a synthesized PHA chain would generate hydrophobic aggregation near the polymer release site. Hydrophobic aggregation of the PHA chain might be stuck and block the release site, thereby preventing effective polymerization by PhaC_Ac (Fig. 8a). In the presence of PhaP, the PhaP molecules adsorb to the PHA chain, which increases PHA's hydrophilicity, thus leading to prevention of PHA chain aggregation (Fig. 8b). In fact, the specific activity of the prepolymerization PhaC_Ac was lower than that of the prepolymerization PhaC_Re (Table 1), which was most likely due to inefficient polymer release of the prepolymerization PhaC_Ac. To further verify the validity of this model, detailed structural information of PhaCs and PhaPs is necessary, which is currently unresolved.

FIG 8.

Proposed model of PhaP function for assisting in the initial stage of PHA polymerization by PhaC_Ac. PhaC_Ac and PhaP are illustrated as dimer and trimer forms, respectively. (a) PhaC_Ac alone. (b) PhaC_Ac with PhaPs. See text for details.

In conclusion, this study demonstrated that PhaPs could function as activators of PhaC_Ac, in addition to controlling the surface properties of PHA granules. PhaC_Ac was activated in vitro by the presence of PhaPs both in the prepolymerization state and the polymer-elongating state. The PhaP_Ac-activated PhaC_Ac exhibited a substrate preference that was slightly shifted toward the C₆ substrate. In contrast, the activities of prepolymerization PhaC_Re and PhaC_Da were decreased by PhaP_Ac, whereas the activity of polymer-elongating PhaC_Re was not affected. PhaP_Ac expression in vivo increased P(3HB) accumulation 2.3-fold and 1.2-fold in the PhaC_Ac strain and PhaC_Re strain, respectively, compared with its accumulation in the corresponding PhaP_Ac-free strains. The enhanced P(3HB) accumulation may be attributed to the role that PhaP_Ac plays in assisting the initial stage of PHA polymerization by PhaC_Ac and its chaperone-like role in PhaC_Re folding. These observations provide new insights into the functions of PhaPs and highlight the importance of PhaPs for effective PHA production.