INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Pseudomonas oleovorans is able to accumulate medium-chain-length poly[(R)-3-hydroxyalkanoates] (mcl-PHAs) as intracellular carbon and energy storage compounds when grown with alkanes (9, 24), mcl-alkanoates (3, 21), n-alkanols (16), or many derivatives thereof (27, 38). mcl-PHAs are hydrophobic polyesters and are attracting increasing interest for their possible application as biologically produced and inherently biodegradable substitutes for conventional plastic materials (6, 8, 10, 20, 32). Particularly, when subjected to nitrogen limitation in batch (22, 24) and chemostat culture (17, 33, 34), P. oleovorans has been reported to accumulate high cellular mcl-PHA contents.

The nature and availability of essential nutrients are important parameters in determining the extent of storage compound accumulation in P. oleovorans (23, 24). Generally, the production of microbial cell mass is limited by the restricted availability of a particular nutrient. In batch culture, the exhaustion of a specific nutrient terminates the exponential-growth phase, whereas in chemostat culture the biomass concentration is usually controlled by the permanent limitation of a single defined nutrient, typically the carbon and energy source. However, several authors have shown that two or more nutrients can simultaneously limit the production of biomass (for a summary, see reference 14). This phenomenon has also been observed for P. oleovorans, where we have found that under defined chemostat steady-state cultivation conditions, the biomass concentration and the cell composition were influenced by simultaneous limitations of the available carbon and nitrogen (12).

All of the previous reports have emphasized that dual-nutrient-limited growth depends on the ability of a microorganism to adjust its cellular composition to different degrees of nutrient limitation. If a microorganism is very flexible in this respect, an extended dual-nutrient-limited growth regime between two single-nutrient limitations should be observed in continuous culture. It is known (19) that the flexibility in cellular composition increases with decreasing growth rates. Therefore, it was predicted (14) that the range of dual nutrient limitation should also depend on the chemostat dilution rate. In order to test this hypothesis and to investigate the influence of the growth rate on the physiological adaptation and mcl-PHA accumulation pattern of P. oleovorans, cells were cultivated in the chemostat at different dilution rates and at various ratios of octanoate to ammonia in the feed medium.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strain, media, and growth conditions. P. oleovorans ATCC 29347, kept as frozen stock in 15% glycerol-citrate E medium at 80°C, was used for all experiments. Modified E medium (39) contained, per liter, 3.5 g of NaNH₄HPO₄ · 4H₂O, 7.5 g of K₂HPO₄, 3.7 g of KH₂PO₄, and 2.9 g of Na₃ citrate · 2H₂O. The pH was adjusted to 7.1 with 10 mM NaOH. This mineral medium was autoclaved and then supplemented with 1 ml of filter-sterilized MgSO₄ (1 M) and 1 ml of MT trace element stock solution, which contained, per liter, 2.78 g of FeSO₄ · 7H₂O, 1.47 g of CaCl₂ · 2H₂O, 1.98 g of MnCl₂ · 4H₂O, 2.81 g of CoSO₄ · 7H₂O, 0.17 g of CuCl₂ · 2H₂O, and 0.29 g of ZnSO₄ · 7H₂O in 1 M HCl.

Continuous cultivation conditions. We used a 50-ml cell suspension grown overnight at 30°C in medium E in 300-ml shaking flask batch cultures as inocula for chemostat experiments. After an initial start-up batch, the bioreactor was switched to continuous mode. The following medium was used for continuous culture cultivations; 1 g of KH₂HPO₄/liter 0.708 g of (NH₄)₂SO₄/liter (equals 150 mg of N/liter), 50 ppm of silicone antifoam, and variable amounts of sodium octanoate or trisodium citrate dihydrate. After autoclaving, 1 ml of 10 mM FeSO₄ · 7H₂O (in 1 M HCl), 1 ml of 1 M MgSO₄, and 1 ml of continuous culture trace metal (CCTM) stock solution were added to the medium by filter sterilization. CCTM stock solution consisted of 1.47 g of CaCl₂ · 2H₂O/liter, 1.98 g of MnCl₂ · 4H₂O/liter, 2.81 g of CoSO₄ · 7H₂O/liter, 0.17 g of CuCl₂ · 2H₂O/liter, 0.29 g of ZnSO₄ · 7H₂O/liter, and 10 g of EDTA/liter at pH 4. The continuous culture experiments were done in a 3-liter laboratory bioreactor (MBR, Wetzikon, Switzerland) with a working volume of 2 liters equipped with automated pH control, a dissolved oxygen tension electrode, a magnetic coupled stirrer, and dilution rate control by an automatic balance.

Sample preparation. Cell samples were taken in ice-cooled flasks and spun down at 10,000 × g for 15 min at 4°C, and the culture supernatant was analyzed for residual nutrient concentrations. The cell pellet was washed twice with 10 mM MgCl₂, frozen in nanopure water, and lyophilized.

Dry weight of cells. The biomass concentration (measured in dry weight of cells [CDW]; also abbreviated as X) in the culture was measured by filtering an appropriate amount of cell suspension (3 to 8 mg, CDW) through a preweighed 0.2-µm-pore-size polycarbonate filter. The filter was washed once with 10 ml of a 10 mM MgCl₂ solution and dried overnight at 105°C. The weight difference after cooling the filter in the exsiccator over silica gel was measured to calculate the concentration of CDW in the culture. All CDW measurements of chemostat experiments were done in triplicate. The relative error was below 5%. CDW determinations in batch cultures were done only once per sample because of the limiting amount of culture volume. Biomass concentration measurements were double-checked by optical density measurements at 450 nm and CDW determination. The results of the two techniques agreed well.

Analysis of substrates in medium and culture supernatant. Ammonium was measured according to the indophenol method described by Scheiner (37). The detection limit of this method was 0.12 mg of N/liter. The method was linear up to concentrations of 2 mg of N/liter. Sample dilutions were done in nanopure water if necessary. The identity of acetate was verified by an enzymatic standard assay (Boehringer, Mannheim, Germany). Octanoate and acetate were measured by a gas chromatography (GC) method by direct injection of acidified aqueous samples (column, Permabond FFAP-0.35, 25 m by 0.32 mm; Macherey-Nagel, Basel, Switzerland). Butyric acid served as the internal standard, and a calibration curve for sodium octanoate in E medium was used as the external standard. Samples and standards were mixed in a 1:1 ratio with a solution containing 1 g of butyric acid/liter and 15 vol% orthophosphoric acid. The detection limit of this method was 50 mg of octanoic acid and acetic acid/liter, and the calibration curve was linear up to 4 g/liter. Dissolved organic carbon (DOC) in culture supernatants was measured using a TOCOR 2 total organic carbon analyzer (Maihak AG, Hamburg, Germany). For this purpose 10 ml of sample was acidified with 150 µl of concentrated HCl and stripped from CO₂ with nitrogen gas. The linear concentration range was 5 to 100 mg of DOC/liter, and dilutions, if necessary, were made in nanopure water.

Elemental analysis. The elemental cell composition was measured with a Carlo Erba (CHNS) elemental analyzer (model EA1108; Carlo Erba, Milan, Italy).

PHA analysis. The PHA content and monomeric composition of the cells were analyzed by a GC method initially described by Braunegg et al. (4) and adapted for mcl-PHAs by Lageveen et al. (24). Methylbenzoate served as the internal standard, and pure mcl-PHA obtained by chloroform extraction and methanol precipitation was used as a methanolysation standard. This standard PHA was weighed and treated in the same way as the lyophilized cell samples. The identities of the monomers were checked with GC-mass spectrometry (MS) measurements (GC-MS MD 800; Brechbühler, AG, Schlieren, Switzerland).

Gas analysis. The exhaust gas stream of the chemostat experiments was continuously analyzed for its CO₂ (by infrared spectroscopy; IR-Binos; Leybold-Heraeus, Hanau, Germany) and O₂ (paramagnetically) contents with a standard gas analyzer (Oxymat 3; Siemens, Erlangen, Germany).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

P. oleovorans was grown in chemostat culture at five different dilution rates ranging from 0.05 to 0.4 h¹ with a mineral medium and octanoate as the sole source of carbon and energy. The carbon-to-nitrogen ratio of the supplied growth medium (C₀/N₀, given in moles per moles) was adjusted by keeping the concentrations of ammonium (N₀ = 10.7 mM) and all other inorganic nutrients constant and by varying the concentration of octanoate stepwise (by varying the octanoate feed pump flow rate). Up to at least 2 g of dry PHA-free biomass liter¹, this medium was clearly limited by either C or N under the experimental conditions used (42). Culture parameters, such as residual concentrations of octanoate and ammonium, biomass produced, and cellular PHA contents were assessed under steady-state conditions (defined as constant concentrations of residual substrates and biomass concentrations).

Three distinct nutritional growth limitations. When the C₀/N₀ ratio of the inflowing growth medium was increased from low to high values, a typical pattern was observed at all dilution rates tested (Fig. 1): in addition to pure carbon limitation at low C₀/N₀ ratios in the medium feed with unutilized ammonium in the culture supernatant and pure nitrogen limitation with residual octanoate at high C₀/N₀ ratios, an intermediate zone of dual nutrient limitation was observed. Under these conditions both carbon and nitrogen were used to completion, and their concentrations in the culture supernatant were below the detection limit. The data clearly demonstrate that the width and location of the dual-nutrient-limited growth zone (Fig. 1) were functions of the growth rate. At a dilution rate of 0.4 h¹, the width of the zone was only 1.6 C₀/N₀ units, whereas at 0.1 h¹, the zone extended over 9.8 C₀/N₀ units. Both the lower and the upper boundaries of the dual-nutrient-limited growth zones shifted to higher C₀/N₀ values as the growth rate decreased.

View larger version (34K): [in this window] [in a new window]   FIG. 1.   Growth and PHA content of P. oleovorans in chemostat culture at different dilution rates, as a function of the carbon-to-nitrogen ratio of the medium feed (C0/N0, in moles per moles) with octanoate as the sole source of carbon and energy and ammonium as the only source of nitrogen. The concentrations of nitrogen and all other inorganic nutrients in the feed were kept constant (N0 = 10.7 mM), whereas the concentration of octanoate was varied. The zone of dual-nutrient-limited growth (shaded) was calculated from the slopes of residual nitrogen (lower boundary) and octanoate (upper boundary). The upper boundary at a D of 0.05 h1 could not be determined due to by-product formation and unstable growth (see the text).

Determination of the boundaries of dual-nutrient-limited growth. The boundaries of dual-nutrient-limited growth were estimated by two different procedures.

1

View larger version (17K): [in this window] [in a new window]   FIG. 2.   Elemental yield coefficients for nitrogen and carbon determined for P. olevorans during growth in the chemostat with octanoate and ammonium under either nitrogen- or carbon-limited conditions, as a function of the dilution rate. Symbols: , YX/N for carbon-limited growth; , YX/N for nitrogen-limited growth; , YX/C for carbon-limited growth; , YX/C for nitrogen-limited growth; , YX/N of PHA-free biomass for carbon-limited growth; , YX/N of PHA-free biomass for nitrogen-limited growth.

View larger version (46K): [in this window] [in a new window]   FIG. 3.   Extension of the dual (carbon and nitrogen)-nutrient-limited growth zone for the cultivation of P. oleovorans with octanoate and ammonium as a function of the dilution rate and the C0/N0 ratio (in moles per moles) of the medium feed. Measured boundaries were obtained from the slopes of either residual nitrogen (lower boundary) or octanoate (upper boundary) versus the C0/N0 ratio in the medium feed. Alternatively, the boundaries from the carbon and nitrogen yield coefficients were calculated according to the method of Egli and Quayle (13). The growth yield coefficients for the maximum growth rate (µmax) were determined in batch culture as described in reference 12 and refer to the non-PHA biomass produced during exponential growth. Lower boundary: , calculated from yield coefficients; , determined from residual nitrogen concentration pattern. Upper boundary: , calculated from yield coefficients; , determined from residual octanoate concentration pattern.

Elemental cell composition. The elemental cell composition with respect to carbon and nitrogen was strongly dependent on the cellular PHA content (Table 1 and Fig. 4). The C/N ratio of the biomass was constant under single-nutrient-limited growth conditions and increased with increasing cellular PHA contents. In contrast, the calculated C/N ratio of the non-PHA biomass remained almost constant. This indicates that some of the surplus octanoate supplied to the culture in the dual-nutrient-limited growth zone was accumulated as mcl-PHA, and accumulation of this storage compound was the most significant factor of the change in the cellular composition with respect to carbon and nitrogen.

View larger version (22K): [in this window] [in a new window]   FIG. 4.   Cellular C/N ratio of P. oleovorans grown with octanoate at different dilution rates as a function of the mcl-PHA content. , C/N ratio of whole freeze-dried cells; , C/N ratio of the non-PHA biomass.

mcl-PHA accumulation as a function of the C₀/N₀ ratio and dilution rate. PHA accumulation was detected in cells of P. oleovorans under all growth conditions tested (Fig. 5). Low cellular PHA contents were measured even during carbon-limited growth at low C₀/N₀ ratios. The cellular PHA content under carbon-limited growth conditions was a function of the dilution rate. It was below 2 wt% at a D of 0.05 h¹, exhibited a maximum of more than 10 wt% at a growth rate of 0.3 h¹, and decreased again to 7.6 wt% at 0.4 h¹. For all dilution rates, the cellular PHA content increased within the dual-nutrient-limited growth zone and remained at the maximum level under pure nitrogen limitation (Fig. 5; compare also Fig. 1). For cells grown under distinct nitrogen limitation, the mcl-PHA content was inversely proportional to the growth rate and decreased from more than 56% at a growth rate of 0.1 h¹ to less than 21% at 0.4 h¹.

View larger version (16K): [in this window] [in a new window]   FIG. 5.   Cellular mcl-PHA contents of P. oleovorans grown in continuous culture with octanoate and ammonium, as a function of the growth-limiting nutrient at different dilution rates. PHA content was measured under carbon ()- and nitrogen ()-limited growth conditions.

Composition of the accumulated polymer. The monomeric composition of the accumulated mcl-PHA was independent of the dilution rate and consisted of 88 mol% 3-hydroxy-octanoic acid (3OH-8) and 12 mol% 3-hydroxy-hexanoic acid (3OH-6). Only when the cellular PHA content was below 3 wt% was no 3OH-6 detected by GC analysis. However, this was possibly due to analytical limitations of the method and not to the total absence of 3OH-6 in the polymer itself.

By-product formation under nitrogen-limited growth conditions at low dilution rates. At dilution rates equal to or lower than 0.1 h¹ and C₀/N₀ ratios in the feed medium that were close to or above the upper boundary of dual nutrient limitation, significant by-product formation was detected. This excretion of by-products always went along with a visible browning of the culture supernatant and an exceptionally low Y_X/C. Although no distinct absorption maxima were observed, UV/visible spectrometry of the culture supernatant showed that the excreted products resulted in an increase in absorption at wavelengths below 450 nm. Size exclusion chromatography indicated that the major fraction consisted of by-products with molecular masses below 250 Da. Acetate was identified as the main by-product by ion chromatography and enzymatic assay. However, because the absolute amount of excreted carbonaceous compounds is unknown, the fraction of acetate can only be estimated to be approximately 60% of the excreted carbon. When the chemostat dilution rate was increased above 0.1 h¹ or the C₀N₀ ratio in the medium feed was decreased below a critical value, the brown color disappeared according to a washout curve. Due to this phenomenon, no purely nitrogen-limited steady-state conditions could be achieved at a dilution rate of 0.05 h¹.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Dual-nutrient-limited growth zones and their boundaries. The cultivation of P. oleovorans in chemostat culture with octanoate and ammonium at different C₀/N₀ ratios of the medium feed led to steady-state growth not only under clearly C- or N-limited conditions. In addition, in between the two single-substrate-limited growth regimes, a stable zone of growth was observed where both nutrients were utilized to completion (referred to as a dual-nutrient-limited growth zone).

(1)

(2)

mcl-PHA accumulation under carbon-limited growth conditions. At growth rates approaching the maximum growth rate (µ_max), the cells produced a considerable amount of mcl-PHA under purely carbon-limited growth conditions (Fig. 5). As we have shown elsewhere (12), P. oleovorans also accumulates mcl-PHA in batch culture during unrestricted growth in the exponential phase. Both observations might result from the fact that in cells cultivated with mcl-fatty acids at high growth rates, the intracellular metabolite pools relevant for PHA accumulation are saturated, and overflow metabolism directs -oxidation metabolites into mcl-PHA precursors and polymer. A similar explanation has been put forward for the growth-associated poly (hydroxybutyric acid) (PHB) accumulation in Methylobacterium rhodesianum (1), where a postulated metabolic bottleneck diverted intermediates into PHB formation. Page and Knosp (29) explained the PHB accumulation during unrestricted growth of Azotobacter vinelandii strain UWD by its deficient capacity to oxidize NADH and the resulting high ratio of NADH to NAD⁺. This excess of reducing power in the cell feeds back on the tricarboxylic acid (TCA) cycle, which in turn promotes the condensation of acetoacetyl coenzyme A (CoA) that allows PHB synthesis as an alternative electron sink (see also reference 30). In both cases PHB was formed as an overflow product of central metabolic pathway intermediates. Several other authors detected growth-associated PHB or mcl-PHA accumulation in the exponential phase of batch cultures in various microorganisms (2, 4, 22, 26, 41).

mcl-PHA accumulation under nitrogen limitation. When cells were subjected to pure nitrogen limitation, the extent of mcl-PHA stored increased with decreasing growth rates. Although the absolute PHA contents reported differed slightly, the same observation has been made previously for P. oleovorans growing under conditions of carbon excess at different dilution rates in a two-liquid-phase chemostat with octane (33). Similar observations have also been reported by Ramsay et al. (35), who cultivated P. oleovorans with octanoate at different dilution rates at a fixed C₀/N₀ ratio of 14.7 (this C₀/N₀ ratio was calculated from the experimental data reported by these authors).

1

1

Dual nutrient limitation as a tool for growth medium optimization. The phenomenon of growth under dual-nutrient-limited conditions at a constant dilution rate has been described previously (7, 11, 13, 15, 25, 28, 36, 40). It has also been conjectured that the zone of multiple-nutrient limitation should be a function of the growth rate (13, 14). However, only one report is known which actually presented some experimental evidence for this hypothesis (28). Here, we conducted the first investigation focusing on the relation between the growth rate and the extension of the dual-nutrient-limited growth zone in combination with the accumulation of a nutrient storage compound.

1