INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Polyhydroxyalkanoic acid (PHA) is accumulated in bacterial cells from many types of carbon sources and it can be used as energy source (3, 25) if the PHA is degraded under certain conditions (10, 39). PHA composition depends on the PHA synthases present, the carbon sources and the route by which the supplied carbon sources are metabolized (25). Medium-chain-length (MCL) PHA-producing bacteria (specifically, the Pseudomonas spp. belonging to rRNA group I) can be used for the production of PHA with functional groups, such as phenyl, phenoxy, olefins, halogens, esters, etc. (25), in its side chains. In particular, the incorporation of carbon-carbon double (15, 22, 24, 27, 37) or triple (21) bonds into the side chains of PHA may be important because a large number of applications in pharmacology, agricultural science, etc., are expected after some modification of the olefins by attaching biological active molecules or cross-linking the multiple bonds by -irradiation (4), UV light (21), or epoxidation (5, 38). A usual preparation of the modifiable PHA requires a cofeeding of expensive alkenoic acids or alkenes. The fatty acids with more than six carbon atoms are degraded via the -oxidation pathway. Intermediates from the -oxidation cycle can be converted to (R)-3-hydroxyacyl-coenzyme A (CoA) by a hydratase, epimerase, or reductase activity, and (R)-3-hydroxyacyl-CoA is utilized as the substrate of the PHA synthase to polymerize them (25). The introduction of functional groups in the side chains is thus possible by using the fatty acids with the corresponding functional groups as the substrates.

Another type of PHA with unsaturated carbon-carbon double bonds in the side chains is produced by several MCL-PHA-producing bacteria such as Pseudomonas citronellolis (9), P. aeruginosa, P. putida KT2442 (12, 18, 19, 20), and Pseudomonas sp. strain NCIMB 40135 (16), grown on unrelated carbon sources such as fructose, glucose, acetate, butyrate, gluconate, -diols, -dicarboxylic acids, etc. The unsaturated monomers identified thus far include 3-hydroxy-cis-5-dodecenoate (C_12:1) and 3-hydroxy-cis-7-tetradecenoate (C_14:1), which usually constitute minor monomers in the PHA in addition to the major constituents 3-hydroxyoctanoic acid and 3-hydroxydecanoic acid (3HD) (9, 12, 18, 19, 20). However, in most cases reported thus far, the overall contents of the two unsaturated monomers were <10 mol%. These unsaturated monomers are known to derive from the intermediates of the fatty acid de novo synthesis pathway (14, 17, 25). The intermediates are present in a form of acyl carrier protein (ACP). It was suggested that the substrate of MCL-PHA synthase was (R)-3-hydroxyacyl-CoA in pseudomonads. Thus, to serve as a substrate for the PHA synthase, (R)-3-hydroxyacyl-ACP must be converted to the corresponding CoA derivative. Recently, it was found that 3-hydroxyacyl-ACP:CoA transacylase phaG plays the role in linking the two pathways, fatty acid synthesis and PHA synthesis (14, 17, 25, 31).

PHA synthesis-related inhibitors can be used to find the metabolic pathway from which precursors for PHA synthesis (18) are supplied, as well as to channel intermediates of a pathway specific to PHA synthesis (14, 29). Only two inhibitors, acrylic acid and cerulenin, have been applied in literature (14, 18, 29). Acrylic acid has been employed to study the relationship between the -oxidation cycle and PHA synthesis. It is known to inhibit acyl-CoA synthase and 3-ketothiolase in gram-negative bacteria (18, 29). It was reported that in a wild-type strain such as P. putida KT2442, PHA accumulation from fatty acids could be inhibited in the presence of acrylic acid (18). However, in recombinant strains such as Escherichia coli (fadR) (29) and P. fragi (14) harboring a PHA synthase gene, PHA accumulation was enhanced or induced. Cerulenin has been used as an inhibitor for PHA synthesis from saccharides because it inhibits fatty acid synthesis (18, 28). However, cerulenin also inhibits cell growth strongly at the same level of concentration as in PHA synthesis. Thus, it was necessary to find a specific inhibitor to inhibit only PHA synthesis and not cell growth. This could be very useful for an efficient pathway routing for the preparation of a specially designed PHA with the cells in active growth. In this study, a cheaper and more potent inhibitor, 2-bromooctanoic acid, was found to very specifically inhibit PHA synthesis of the bacterium from saccharides without any influence on the cell growth. 4-Pentenoic acid was also found to inhibit PHA synthesis, but it did so less specifically than 2-bromooctanoic acid.

We were interested in screening bacteria to produce PHA with high levels of those unsaturated monomer units from cheaper carbon sources such as saccharides. Recently, we successfully isolated a bacterial strain that is capable of producing MCL-PHA containing the two unsaturated monomers at up to 40 mol% from fructose. This characteristic strain was identified as a psychrotrophic P. fluorescens. For the three inhibitors, acrylic acid, 2-bromooctanoic acid, and 4-pentenoic acid, the inhibitory effect on the PHA synthesis of the isolated bacterium was investigated over a wide range of concentrations in order to systematically study the relation between the precursor supplying routes and PHA formation in the bacterium. In addition, several saccharides and monocarboxylic acids including C₂ (acetate) to C₁₈ (stearic acid) were tested for their utilization by the bacterium to see what types of PHA are synthesized in the psychrotrophic strain.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Isolation and characterization of strain BM07. The sample obtained from activated sludge in a municipal wastewater treatment plant in Chinju, Korea, was suspended with sterilized water and inoculated to an M1 mineral salts medium (9) containing 0.05% ammonium sulfate and 1% octanoic acid. After 2 days of incubation at 30°C and 200 rpm for 48 h, the culture was diluted 10⁵ fold and 100 µl of the dilution was inoculated onto octanoate M1 mineral agar plates and cultivated for 24 h at 30°C. Several single colonies were picked out according to their opacity, usually caused by PHA synthesis in cells (34), and then purified by a series of spreading steps onto octanoate agar plates. The colony with the highest degree of opacity and the largest size was picked out.

16S rRNA gene sequencing. The strain was also identified by using 16S rRNA gene sequence homology. Genomic DNA was obtained from bacteria grown overnight at 30°C in 5 ml of Luria-Bertani medium. The Marmur procedure (26) was employed for the isolation of the genomic DNA.

Culture media and PHA accumulation in P. fluorescens BM07. Nutrient-rich (NR) medium was used in the seeding, maintenance, and storage of the isolated strain and contained 1% yeast extract, 1.5% nutrient broth, and 1% ammonium sulfate. A modified M1 mineral salts medium of the same composition as that reported earlier (9) was used as the PHA synthesis medium. The culture (5 ml) grown in NR medium at 30°C at 180 rpm for 12 h was transferred to 500 ml of M1 mineral salts medium containing an appropriate amount of a carbon source and 1.0 g of ammonium sulfate/liter in a 2-liter flask and cultivated long enough to grow maximally. The cells were then harvested, washed with methanol, and dried under a vacuum at room temperature. In the cultivation with insoluble substrates such as C₁₄, C₁₆, and C₁₈ carboxylic acid as a carbon source, a two-phase slurry cultivation technique reported earlier (35) was used for an efficient cultivation of the strain.

Polyester isolation and characterization. Polyesters were extracted from an appropriate amount of cells, which had been dried overnight at 50°C under a vacuum, with hot chloroform in a Pyrex Soxhlet apparatus for 6 h. After concentration, the solvent extract was precipitated in rapidly stirred cold methanol. The isolated polymers were dried overnight under a vacuum at ambient temperature and then weighed. Quantitative determination of the monomer units in the polymers was performed by GC as described above. The standardization of each GC peak was made against the PHA of known structure characterized by quantitative nuclear magnetic resonance (NMR) analyses (34, 35).

Transmission electron microscopy. The washed cells were doubly fixed with 2% glutaraldehyde and 1% osmium tetroxide. Ultrathin sectioning was performed by using an LKB-Ultratome with a diamond knife. These sections were then collected on a copper grid coated with a Formvar-carbon film and were poststained with lead citrate and uranyl acetate (34). Electron micrographs were obtained with a Hitachi H-600 electron microscope (Tokyo, Japan) under an acceleration voltage of 75 kV.

Inhibition of PHA synthesis and cell growth. Among the several compounds tested, the three compounds, acrylic acid, 2-bromooctanoic acid, and 4-pentenoic acid were found to be potent inhibitors for PHA synthesis in P. fluorescens BM07. After cultivation of the cells in NR medium at 30°C for 24 h, the cells (2.79 g/liter in dry biomass) were transferred to a modified M1 PHA synthesis medium containing 70 mM fructose or 40 mM octanoate and cultivated at 30°C for 48 h in the presence of an appropriate amount of each inhibitor. The medium contained 1.0 g of ammonium sulfate per liter. The NR medium-grown cells did not contain PHA. Almost an equal amount of NR medium-grown cells (2.79 g/liter in dry cell weight) was added to each culture medium containing a different level of inhibitor. The content of PHA and its monomer composition were measured by GC analysis of the inhibitor-treated cells. Minimum triplicate experiments were carried out and statistically averaged. The standard deviations for each determination were within 5%. Five batches of noninhibitor cells were grown on 70 mM fructose and 40 mM octanoate, respectively, since the control experiments for both carbon sources and their biomasses were measured and averaged for each carbon source. The average values were determined to be 5.281 and 3.665 g/liter for fructose- and octanoate-grown cells, respectively, and were used as the control biomass in the calculation of the percent inhibition. For the three inhibitors, no systematic and significant change in the monomer ratio of the PHA recovered was observed compared to the PHA synthesized in the absence of inhibitor. Irrespective of the level of each inhibitor, a maximum ± 3% (2-bromooctanoic acid) to ± 7% (acrylic acid and 4-pentenoic acid) of nonsystematic GC signal variation was observed for each GC peak. So, the calculation of the percent inhibition was based on the change of the signal intensities of two strong GC peaks, C_12:1 peak for fructose-grown cells and 3-hydroxyoctanoate (3-HO) peak for octanoate-grown cells.

Assay of the inhibitors remaining in media. 2-Bromooctanoic acid that remained in a culture medium was extracted with chloroform, and the chloroform extract was reacted in a sulfuric acid-methanol mixture. The methyl ester in the organic layer was analyzed by using GC. The concentration of acrylic acid and 4-pentenoic acid that remained in media was analyzed in a rather different way. Five milliliters of the culture medium was saturated with magnesium sulfate, and 0.5 ml of 2 M HCl was added to the saturated solution in order to increase the efficiency of transfer of the remaining inhibitor into an organic phase. After the mixture was vortex mixed, the free acid was extracted by adding 2 ml of chloroform. The chloroform extract was methanolized in a similar manner, and the resulting methyl ester was analyzed by using a Hewlett-Packard 5890A gas chromatograph equipped with a DB-WAXETR column (30 m by 0.53 mm by 0.25 µm; J&W Scientific, Folsom, Calif.). A typical GC run condition is as follows: initial temperature, 60°C for methylacrylate and 50°C for methyl-4-pentenoate; initial time, 2 min; heating rate, 10°C/min; and final temperature, 120°C. The methyl ester of acrylic acid and 4-pentenoic acid was eluted at retention times of 3.2 and 2.8 min, respectively. The same treatment procedure was applied to the samples for the standardization curves of the two inhibitors.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Characterization of the isolated strain. Table 1 shows the morphological, biochemical, and nutritional characteristics for the isolated strain. The strain has three to five polar flagella, and it exhibited motility. The strain is able to grow at 4°C. Activities of cytchrome oxidase and arginine dihydrolase were present. Denitrification was observed. Esculin was not hydrolyzed, but gelatin was hydrolyzed. Glucose, L-arabinose, and D-mannose were utilized for growth. Phenylacetate was not utilized. Thus, from the test result obtained by using the API identification program (ID-32-GN strip test) in accordance with standard methods (23), the isolate was identified as a strain of P. fluorescens (%ID = 99.5) and named P. fluorescens BM07. The strain was also identified by 16S rRNA gene sequence homology (98%) as P. fluorescens. An almost complete 16S rDNA sequence (1,469 nucleotides) was determined and compared with known 16S rDNA sequences.

PHA synthesis from unrelated carbon sources. Among the seven saccharides tested, fructose was the best carbon source for both cell growth and PHA production. The growth curve for the cells grown at 30°C on 70 mM fructose in a mineral salts medium containing 1.0 g of ammonium sulfate/liter (molar C/N ratio of ~10) is shown in Fig. 1. Cell growth and PHA accumulation occurred in a growth-associated fashion without any time lag. The dry cell weight was 4 g/liter in a 500-ml batch culture. The yield of PHA synthesis was 25% (wt/wt) dry cells.

View larger version (30K): [in this window] [in a new window]   FIG. 1.   Time course profiles for growth-associated PHA synthesis in P. fluorescens BM07 grown in PHA synthesis mineral salts medium containing 70 mM fructose and 1.0 g of ammonium sulfate/liter at 30°C. PHA formation was monitored by determining the content of PHA in cells by using a sulfuric acid-methanol reaction mixture and GC determination of the resulting 3-hydroxy-methylesters.

View larger version (21K): [in this window] [in a new window]   FIG. 2.   500 MHz 1H-NMR spectrum (A) and 125 MHz 13C-NMR spectrum (B) of PHA synthesized by P. fluorescens BM07 cells grown on 70 mM fructose at 25°C. For the inserted doublet peaks (B), the downfield peaks are associated with saturated monomer units and the upfield ones are due to unsaturated monomer units (C12:1, C14:1, etc.). Thus, from the peak area ratios of the doublets, the relative amount of total unsaturated monomers in PHA can be calculated. The calculated percentage of unsaturated monomer units was 38%, a result that agreed well with the percentage determined by GC (37%). See the text for details.

PHA synthesis from monocarboxylic acids. The lower aliphatic acids (acetic, propionic, butyric, valeric, and capric acids) supported only cell growth and not PHA accumulation. The dry cell mass was ca. 2 g/liter for the five carbon sources. For the longer carbon sources, heptanoic acid (C₇) to hexadecanoic acid (C₁₆), the monomer composition of PHA depends on the length of substrate molecules (Table 2), similar to what is observed in other pseudomonads belonging to rRNA homology group I (23). For the three carbon sources, C₇, C₈, and C₉ acids, the monomer unit that has the same number of carbon atoms as the carbon source used was a major constituent in the PHA produced. Decanoic acid (C₁₀)-grown cells produced the PHA whose major monomer was 3HO, shorter than the substrate by one ethylene unit. For the carboxylic acids longer than C₁₀, 3HO and 3HD almost equally constituted major monomers. In the PHA from the cells grown on C₁₀ to C₁₆, the longer monomer 3HDD was incorporated into the polymer at a significant level (17 to 23 mol%). In the cells grown with C₁₄ and C₁₆, the relatively long 3-hydroxyacid monomers, 3-hydroxytetradecanoic acid and 3HHD, were also polymerized without being degraded via -oxidation. The incorporation levels were 7 and 5 mol% for 3-hydroxytetradecanoic acid and 3HHD, respectively. It seems highly probable that P. fluorescens BM07 could incorporate the monomer with a protected reactive endgroup attached to its terminal into the polymer chain without being modified; the resulting PHA could be used for the production of functional PHA.

View larger version (75K): [in this window] [in a new window]   FIG. 3.   Electron microscopic photographs of PHA inclusion granules in P. fluorescens BM07 cells cultivated with C16 (palmitic acid) (top panel) for 50 h and C18 (stearic acid) (bottom panel) for 82 h at 30°C. The palmitate-grown cells contained 27% (wt/wt) of dry cells of PHA. The palmitate-grown cells clearly showed the presence of large PHA inclusion granules. However, the stearic acid-grown cells revealed tiny electron-transparent inclusions. The tiny inclusions are considered to probably represent the granules containing free stearic acid (5% [wt/wt] dry cells).

Thermal properties of PHA. Table 2 also shows thermal transition data for PHA synthesized in P. fluorescens BM07. The precipitated and dried polymer samples were fully crystallized at an ambient temperature. The PHA prepared from fructose had a lower glass transition temperature (T_g) than that from octanoic acid by 17°C. It did not exhibit a melt transition ca. 50°C, whereas the PHA from carboxylic acid-grown cells did. The absence of the melt may be caused by the cis-type olefinic bonds in the side chains (e.g., C_12:1 and C_14:1 monomers) perturbing side chain packing for crystallization (27). These planar olefin bonds, however, may contribute to a lowering of T_g, resulting from an increased side chain mobility. A small endothermic peak observed at 0°C for the first-run sample of the PHA from fructose-grown cells appeared again during the reheating process after the first run scanned up to 150°C. Thus, the melting peak at 0°C may be ascribed to a structural characteristic of the PHA and not to an artifact. The polyesters prepared from C₁₄ and C₁₆ substrates have a higher percentage of longer monomers, 3HD and 3HDD (total, 50 mol%), than that from octanoic acid. However, the melting point of the longer side chain PHA is not significantly different from that of the 3-HO-dominant PHA, which means that the uniformity of side chain length may not be a principal factor in determining the crystallinity of MCL PHA.

Inhibition of PHA synthesis and cell growth by inhibitors. The dependence of inhibitor concentration on PHA synthesis and cell growth was investigated for all three inhibitors: acrylic acid, 2-bromooctanoic acid, and 4-pentenoic acid (Fig. 4, 5, and 6). The addition of an inhibitor to a cell growth medium could affect the growth and PHA accumulation kinetics of P. fluorescens BM07. Therefore, in order to compare the inhibitory effects of the three inhibitors in terms of strength and specificity, it is essential to determine the sampling time for each culture exhibiting a steady-state maximum cell growth and PHA synthesis in the presence of inhibitor. As shown in Fig. 1, the increasing patterns of the time course profiles of optical density (OD), cell yield, and PHA content are in parallel with one another. Furthermore, it was reported that an increase in PHA content linearly and significantly increased the OD of a nitrogen-free culture medium in which R. eutropha H16 was grown (33). Thus, a steady-state plateau OD value must be attributed to both maximal cell growth and PHA accumulation. Thus, we investigated the time courses for the growth of NR medium-grown cells on fructose in the presence of 5 mM acrylic acid, 1 mM 2-bromooctanoic acid, or 3 mM 4-pentenoic acid and on octanoate in the presence of 3 mM acrylic acid, 3 mM 2-bromooctanoic acid, or 3 mM 4-pentenoic acid (data not shown). For fructose-grown cells, none of the three inhibitors shifted the time to reach the plateau OD compared to the control growth experiment without inhibitor (the four OD profiles almost exactly overlapped one another). For octanoate-grown cells, the OD profiles for the three inhibitor-treated cells were shifted by 10 h compared to the inhibitor-free control experiment, due to the almost equally increased induction periods. However, the time leading to the plateau OD was in a similar range 45 to 50 h. Thus, both the maximal cell growth and PHA synthesis were considered to occur within 42 to 48 h of cultivation for all experiments. Therefore, the sampling was made after 48 h of cultivation for all inhibition experiments.

View larger version (21K): [in this window] [in a new window]   FIG. 4.   Inhibition of PHA synthesis () and cell growth () by acrylic acid when P. fluorescens BM07 was grown on 40 mM octanoic acid (open symbols) or 70 mM fructose (solid symbols) at 30°C. NR medium-grown cells were transferred to PHA synthesis medium containing the indicated level of acrylic acid and were grown for 48 h. Cell growth was determined by measuring the dry cell weight (DCW). The calculation of the percent inhibition of PHA synthesis was based on the change of the signal intensities of two strong GC peaks, a C12:1 peak for fructose-grown cells and a 3-HO peak for octanoate-grown cells. Minimum triplicate experiments were carried out and statistically averaged. The percent inhibitions for PHA accumulation and cell growth are defined in Materials and Methods. The limiting value corresponding to the theoretically allowed maximum percent inhibition of cell growth was calculated to be 35% for fructose-grown ( ·  ·  · ) and 10% for octanoate-grown ( · · · · · · · · ) cells. Thus, the calculated apparent percent inhibition (virtually not normalized) of cell growth exceeding the limiting value was considered to be 100% inhibition of cell growth.

View larger version (22K): [in this window] [in a new window]   FIG. 5.   Inhibition of PHA synthesis () and cell growth () by 2-bromooctanoic acid when P. fluorescens BM07 was grown on 40 mM octanoic (open symbols) acid or 70 mM fructose (solid symbols). The limiting value corresponding to the theoretically allowed maximum percent inhibition of cell growth was calculated to be 35% for fructose-grown ( ·  ·  · ) and 10% for octanoate-grown ( · · · · · · · · ) cells. Thus, the calculated apparent percent inhibition (virtually not normalized) of cell growth exceeding the limiting value was considered to be 100% inhibition of cell growth. See the legend in Fig. 4 for the other details.

View larger version (20K): [in this window] [in a new window]   FIG. 6.   Inhibition of PHA synthesis () and cell growth () by 4-pentenoic acid when P. fluorescens BM07 was grown on 40 mM octanoic acid (open symbols) or 70 mM fructose (solid symbols). The limiting value corresponding to the theoretically allowed maximum percent inhibition of cell growth was calculated to be 35% for fructose-grown ( ·  ·  · ) and 10% for octanoate-grown ( · · · · · · · · ) cells. Thus, the calculated apparent percent inhibition (virtually not normalized) of cell growth exceeding the limiting value was considered to be 100% inhibition of cell growth. See the legend in Fig. 4 for the other details.

Metabolism of three inhibitors in PHA synthesis medium. Analysis of the metabolic fates of the three inhibitors must be helpful for understanding the above inhibition profiles in terms of their specificities. Thus, the catabolism of the three inhibitors in P. fluorescens BM07 was investigated (Table 3). NR medium-grown cells were cultured in PHA synthesis medium with an appropriate level of an inhibitor. The medium contained 70 mM fructose or 40 mM octanoate as a carbon source and 1 g of ammonium sulfate/liter. The inhibitor remaining in medium was analyzed by GC. When added at millimolar concentrations, both acrylic acid and 4-pentenoic acid completely disappeared after 48 h of cultivation. On the contrary, 2-bromooctanoic acid was catabolized very little in both fructose and octanoic acid medium. In addition, the disappearance of 2-bromooctanoic acid is a function of the initial fed concentration, a result strongly suggestive of highly specific inhibition by one of its metabolites. However, none of the three inhibitors were utilized by P. fluorescens BM07 when they were fed as a sole carbon source at millimolar levels (1 to 5 mM). Acrylic acid is known to be metabolized to CO₂ and acetyl-CoA via a secondary pathway of propionic acid catabolism in mouse tissues (6). Thus, 3-hydroxypropionic acid could be a probable inhibitor. However, 3-hydroxypropionic acid did not affect PHA accumulation from both octanoic acid and fructose at the level of 5 mM. Its ineffectiveness may be due to no transportation of the molecule into the cell because P. fluorescens BM07 did not grow on 40 mM 3-hydroxypropionic acid as a sole carbon source.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

High production of unsaturated 3-hydroxy acids from saccharides. As far as we know, this is the first report of a bacterial strain capable of producing such a high amount of the unsaturated monomers 3-hydroxy-cis-5-dodecenoate (C_12:1) and 3-hydroxy-cis-7-tetradecenoate (C_14:1) from unrelated carbon sources such as saccharides. Hence, it is suggested that this strain can be used as a producer of the two unsaturated fatty acids, which presumably are useful for pharmaceutical applications, from cheap saccharides. We do not know at present why the bacterium produces such high levels of these unsaturated 3-hydroxy acids from saccharides. However, in light of the good growth at 5°C, the high level of production of the unsaturated 3-hydroxy acids may be related to a means for survival in cold environments. This idea was supported by our preliminary experimental finding that a lowering of the cultivation temperature led to an increase in the level of unsaturated derivatives in PHA, as well as in membrane lipids (data not shown).

Acrylic acid and 4-pentenoic acid are multiple-site PHA synthesis inhibitors. The three inhibitors exhibited different inhibitor concentration dependencies on PHA synthesis. This may imply that their curve shapes strongly reflect their catabolism behavior. In particular, the inhibition curves for acrylic acid and 4-pentenoic acid indicate a probable existence of multiple-site inhibition by one inhibitory species or more than one different species with different K_i values. As shown in Table 3, acrylic acid and 4-pentenoic acid were completely catabolized after 48 h of cultivation. Thus, it is probable that two or more metabolites contribute to their inhibition. Acrylic acid is known to be metabolized to CO₂ and acetyl-CoA via a catabolic pathway of propionic acid in mouse tissues (6). However, since a probable intermediate 3-hydroxypropionic acid was found to be unable to inhibit PHA synthesis, we propose one other intermediate species may inhibit two different enzymes. Thus, the sigmoidal shape of the percent PHA synthesis inhibition curve for acrylic acid (Fig. 4) suggests that acrylic acid, probably in the form of CoA, has at least two target enzymes for inhibiting both PHA accumulation and cell growth. Acrylic acid is known to inhibit the 3-ketoacyl-CoA thiolase, which catalyzes the final step of -oxidation, i.e., the release of acetyl-CoA from 3-ketoacyl-CoA (18). It is also known to inhibit acyl-CoA synthase (18). Such sigmoidal inhibition may indicate a dual inhibition for two different enzymes with different K_i values by acrylic acid. However, in recombinant strains such as E. coli (fadR), acrylic acid was not significantly catabolized (29). Compared to the case of acrylic acid, 4-pentenoic acid showed no such characteristic concentration dependence on PHA synthesis from fructose but rather demonstrated a featureless dependence. 4-Pentenoic acid is known to be metabolized to 2,4-pentadienyl-CoA, 3-keto-4-pentenoyl-CoA, etc. 3-Keto-4-pentenoyl-CoA acts both as a reversible and as an irreversible inhibitor of the 3-ketoacyl-CoA thiolase in rat heart mitochondria (32). 3-Keto-4-pentenoyl-CoA is known as a rather unspecific inhibitor because it inhibits other enzymes such as carnitine acetyltransferase and acetoacetyl-CoA thiolase. (7, 40). The monotonously increasing and broad featureless inhibition curve may thus indicate the presence of at least two or more enzymes interacting with the intermediates derived from 4-pentenoic acid.

4-Pentenoic acid and 2-bromooctanoic acid inhibit the enzyme(s) linking the two pathways, fatty acid synthesis and PHA synthesis. The addition of 4 to 5 mM acrylic acid to fructose medium caused little effect on PHA synthesis and cell growth, but its addition to octanoic acid medium completely inhibited both PHA synthesis and cell growth. This means that most CoA monomers necessary for PHA synthesis in fructose-grown cells may be directly supplied via the route linking fatty acid synthesis and PHA synthesis. It is recently known that the enzyme (R)-3-hydroxylacyl-ACP-CoA transferase converting (R)-3-hydroxylacyl-ACP to (R)-3-hydroxylacyl-CoA plays the role in linking the two pathways (14, 17, 31). 4-Pentenoic acid is known to be a -oxidation inhibitor in animal cells, as described previously. However, 4-pentenoic acid inhibited PHA synthesis strongly in fructose-grown P. fluorescens BM07 cells but only weakly in octanoate-grown cells. This indicates that one of the derived intermediates may act as a strong inhibitor for the enzyme involved in the linkage between fatty acid synthesis and PHA synthesis.

2-Bromooctanoic acid is more potent and specific than cerulenin. Cerulenin inhibits the growth of P. putida KT2442 at 1.34 mM (18). It binds to the active sites of keto-acyl-ACP synthases I and II in the fatty acid synthesis pathway. In glucose-grown cells cerulenin completely inhibited PHA synthesis. Cerulenin also inhibited PHA synthesis by up to 69% in the cells grown on 10 mM octanoate. However, both 4-pentenoic acid and 2-bromooctanoic acid inhibited PHA synthesis by only ~20% in P. fluorescens BM07 grown on octanoic acid. Thus, 2-bromooctanoic acid may inhibit PHA synthesis more specifically than cerulenin. It is therefore suggested that 2-bromooctanoic acid can substitute for the more expensive (2,000 times) and less-specific (cell growth-inhibiting) inhibitor, cerulenin, in the inhibitor study of PHA synthesis.