Tailored Nutrient Feeding Strategies for mcl-PHA Production and Molecular Weight Regulation in Pseudomonas putida

Abstract

Medium-chain-length polyhydroxyalkanoates (mcl-PHAs) are biodegradable microbial polyesters with attractive elastomeric properties and considerable potential as sustainable alternatives to fossil-derived polymers. Their industrial production requires not only efficient biomass and biopolymer productivity but also consistent material quality. In this study, Pseudomonas putida KT2442 was cultivated under different nutrient conditions to evaluate the effects of carbon, nitrogen, phosphorus and magnesium supplementation on biomass growth, poly(3-hydroxyoctanoate) (PHO) accumulation and its properties. Flask-scale experiments identified octanoic acid as an effective precursor for PHO biosynthesis, whereas excessive substrate loadings negatively affected culture performance. Continuous feeding of carbon and nitrogen sources in a fed-batch bioreactor increased biomass concentration to approximately 20 g/L while maintaining a PHO content of about 50% w/w of dry cell weight. Phosphorus limitation did not improve PHO accumulation and instead reduced biomass formation. Continuous magnesium feeding increased biomass concentration to approximately 26 g/L while maintaining intracellular PHO content. The produced PHO exhibited condition-dependent differences in selected molecular weight characteristics, while its thermal transition temperatures showed only small absolute variations. Overall, the proposed nutrient feeding strategies can provide a robust basis for improving process productivity while maintaining consistent biopolymer quality.

1. Introduction

Plastics have become indispensable materials because of their low cost, processability, durability and wide range of properties. However, their dependence on fossil resources and the persistence of plastic waste in the environment have created a strong need for more sustainable polymer production and end-of-life strategies [1,2]. Bio-based and biodegradable polymers are therefore increasingly considered as complementary alternatives to conventional plastics, particularly in applications where material recovery is difficult or where biodegradability and biocompatibility provide added value [3,4,5,6]. Among these materials, polyhydroxyalkanoates (PHAs) represent a versatile family of microbial polyesters, because their chemical composition, molecular properties and final performance can be influenced by the selected strain, substrate and cultivation strategy [3,4,5,6,7,8]. PHAs are intracellular carbon/energy storage compounds accumulated by many bacteria, usually when carbon is available in excess and at least one other nutrient becomes limiting or unbalanced [7,8,9,10]. Depending on their monomer composition and molecular structure, PHAs may exhibit properties ranging from thermoplastic and relatively brittle behaviour to flexible and elastomeric performance [11,12,13,14]. This diversity creates the possibility of tuning biopolymer properties through bioprocess design, allowing the development of PHA materials with property windows suitable for targeted applications [15,16,17,18,19,20].

PHAs are commonly classified into short-chain-length PHAs (scl-PHAs), typically composed of C3–C5 hydroxyalkanoate monomers, and medium-chain-length PHAs (mcl-PHAs), composed mainly of C6-C14 monomers [11,12,13,14,21,22,23,24]. Scl-PHAs, such as poly(3-hydroxybutyrate) (PHB), are usually more crystalline and mechanically stiff, whereas mcl-PHAs are generally less crystalline, have lower glass transition temperatures and show more flexible or elastomeric properties [13,14,21,22,23,24,25]. These differences strongly influence the application profile of each material, with scl-PHAs being more relevant for rigid or semi-rigid applications and mcl-PHAs being attractive for flexible films, coatings, soft packaging, adhesives, etc. [21,22,23,24,25]. Therefore, the evaluation of a PHA production bioprocess should not be limited to biopolymer yield, but should also consider molecular weight, thermal transitions, crystallinity and the potential application window of the produced material [15,16,17,18,19,20,21,22,23,24,25].

The concept of controlling PHA properties through bioprocess conditions has been demonstrated mainly for PHB, where cultivation variables were shown to affect not only biomass formation and biopolymer accumulation but also the molecular weight distribution (MWD) of the recovered biopolymer [15,16,17,18,19,20]. In previous studies, PHA production was treated as an integrated metabolic and polymerization process, allowing molecular properties to be considered controllable quality attributes rather than only as post-fermentation analytical results [16,17,18,19,20]. More recently, statistically optimized PHB cultures have also shown that production performance and biopolymer molecular weight (MW) can be evaluated simultaneously [15]. This process-property correlation is highly relevant for mcl-PHAs, where relatively small changes in monomer composition or molecular weight may significantly affect processability and application performance [21,22,23,24,25].

Among mcl-PHA-producing microorganisms, Pseudomonas putida is one of the most studied and versatile bacterial platforms [26,27,28]. This species can accumulate mcl-PHAs from structurally related carbon sources like fatty acids, as well as from structurally unrelated substrates, including sugars, glycerol, biomass and waste feedstocks [28,29,30,31,32,33,34,35]. When fatty acids are used as substrates, the β-oxidation metabolic pathway plays a central role in determining the monomer composition of the resulting biopolymer [30,35], while octanoic acid is particularly relevant for the production of C8-rich mcl-PHAs; its use, however, requires careful control because fatty acid accumulation can inhibit microbial growth and affect the metabolic balance of the bacteria [30,36,37,38,39]. Beyond P. putida, Pseudomonas oleovorans has also been successfully employed for the production of different mcl-PHAs. Depending on the supplied fatty acid precursor, it has been used to produce poly(3-hydroxyoctanoate) from octanoic acid, poly(3-hydroxynonanoate) from nonanoic acid and unsaturated PHAs from plant-derived unsaturated fatty acids [35]. Notably, the chemical structure of the supplied substrate can direct the monomer composition and properties of the resulting PHA.

Traditionally, PHA accumulation has been associated with excess carbon and limitation of nutrients such as nitrogen, phosphorus, oxygen, sulfur or magnesium [7,8,9,10,40]. This concept has been highly influential, particularly for scl-PHA-producing bacteria, where two-stage processes involving an initial growth phase followed by a nitrogen- or phosphorus-limited accumulation phase have been widely applied [8,9,10,40]. However, the application of this logic to mcl-PHA-producing strains is not always straightforward, because PHA synthesis may remain strongly associated with active growth and carbon flux through fatty acid metabolism [28,36,37,38,39]. Several studies on Pseudomonas sp. have shown that feeding mode, carbon, oxygen and nutrient supplementation can strongly direct the strain’s metabolism, suggesting that the dynamic feeding trajectory may be more important than the simple presence of a nominal nutrient limitation [28,36,37,38,39].

Fed-batch cultivation is therefore a particularly important operating mode for mcl-PHA production, as it allows the gradual supply of inhibitory or poorly soluble carbon sources while maintaining active biomass formation and biopolymer accumulation [28,36,37,38,39]. Tailored feeding strategies have been used to avoid excessive fatty acid concentrations, to reduce substrate inhibition and to sustain high-cell-density cultivation. Previous fed-batch studies with P. putida have primarily established the importance of limiting inhibitory fatty acid accumulation, applying carbon-limited operation or glucose co-feeding, and managing carbon and nitrogen availability to sustain biomass formation and mcl-PHA production [28,31,32,33,34,36,40]. However, phosphorus and particularly magnesium are still generally treated as fixed medium constituents, and their availability or feeding dynamics have not been systematically examined alongside carbon and nitrogen under a common fed-batch framework. Moreover, relatively few studies have linked these nutrient supply strategies to the evolution of PHO molecular weight and thermal properties [15,16,17,18,19,20,21,22,23,24,25]. The novelty of the present work therefore lies in the integrated assessment of carbon, nitrogen, phosphorus and magnesium supply strategies in P. putida KT2442 and in relating the resulting culture performance and PHO productivity to molecular weight evolution and thermal stability.

In the present work, Pseudomonas putida KT2442 was investigated as a mcl-PHA-producing strain under different nutrient feeding conditions, with emphasis on the combined effects of carbon, nitrogen, phosphorus and magnesium supply on microbial growth, PHO accumulation and selected biopolymer properties. Particular emphasis was placed on the comparison between pulse and continuous feeding, the role of phosphorus limitation and the effect of magnesium supplementation during fed-batch cultivation. The overall objective was to develop a bioprocess-oriented understanding of how nutrient feeding strategies influence PHO production performance and whether they affect molecular-weight evolution and thermal properties of the recovered biopolymer. This approach supports the broader development of controlled fed-batch strategies for improving mcl-PHA production performance.

2. Materials and Methods

2.1. Microorganism and Cultivation Medium

Pseudomonas putida KT2442, kindly provided by ADM—BIOPOLIS S.L. (Valencia, Spain), was used throughout this study as an intracellular medium-chain-length polyhydroxyalkanoate (mcl-PHA) and specifically as a poly(3-hydroxyoctanoate) (PHO) producer. The bacterial strain was maintained on nutrient agar plates at 4 °C. For inoculum preparation, cells from a single colony were first cultivated in 100 mL Erlenmeyer flasks containing 25 mL of Luria–Bertani (LB) medium composed of 5 g/L yeast extract, 10 g/L tryptone and 5 g/L NaCl. The flasks were incubated at 30 °C and 150 rpm for approximately 8 h in a GFL 3031 shaking incubator (Berlin, Germany). Subsequently, 5 mL of the culture was transferred to 250 mL Erlenmeyer flasks containing 50 mL of a preculture medium composed of 4.70 g/L (NH₄)₂SO₄, 0.80 g/L MgSO₄·7H₂O, 12.0 g/L Na₂HPO₄·12H₂O, 2.70 g/L KH₂PO₄ and 3.0 g/L nutrient broth (NB). The main experiments were performed in a chemically defined medium, following formulations previously applied to the same strain [31,32]. Unless otherwise stated, the medium contained: 2.5 g/L (NH₄)₂SO₄, 2.03 g/L KH₂PO₄, 9.0 g/L Na₂HPO₄·12H₂O, 0.8 g/L MgSO₄·7H₂O and 10 mL/L trace element solution. Octanoic acid and glucose were used as the main carbon sources at initial concentrations of 25 mM and 20 g/L, respectively, unless otherwise specified. The trace element solution contained, per litre of 1 N HCl solution: 10 g FeSO₄·7H₂O, 3 g CaCl₂·2H₂O, 2.2 g ZnSO₄·H₂O, 0.5 g MnSO₄·4H₂O, 0.3 g H₃BO₃, 0.2 g CoCl₂·6H₂O, 0.15 g Na₂MoO₄·2H₂O, 0.02 g NiCl₂·6H₂O and 1.0 g CuSO₄·5H₂O.

2.2. Fermentation Conditions and Feeding Policies

The flask-scale experiments were conducted as batch cultures in 2 L Erlenmeyer flasks containing an initial working volume of 0.5 L, corresponding to a working-volume-to-flask-capacity ratio of 0.25. The flasks were incubated in a GFL 3033 orbital shaking incubator (Berlin, Germany) at 30 ± 0.1 °C and 210 rpm. The cultures were maintained for approximately 26–28 h and were terminated after reaching the stationary phase. No active pH or dissolved oxygen control was applied at flask scale; oxygen transfer was provided by orbital shaking and gas exchange through the flask headspace. All medium components were supplied at the beginning of the batch cultivation, except in experiment F-N2, in which an additional 0.5 g/L of ammonium sulfate was supplied as a single pulse after 10 h of cultivation. The flask-scale experiments were coded as F-C, F-OA and F-N, corresponding to the carbon-source screening, initial octanoic acid concentration screening, and nitrogen availability and pulse feeding experiments, respectively.

The bioreactor experiments were conducted in fed-batch mode using a 2 L stirred-tank glass bioreactor (BioFlo 110, New Brunswick Scientific, Edison, NJ, USA). The initial working volume was 1 L, corresponding to an initial working-volume-to-nominal-vessel-volume ratio of 0.5. The temperature was maintained at 30 ± 0.1 °C, while the pH was controlled at 7.0 ± 0.05 via the automatic addition of 3 M NaOH or 3 M HCl solutions. The culture was aerated at 1 vvm. Dissolved oxygen was maintained at 25% of air saturation by automatically adjusting the impeller speed between 100 and 1000 rpm. When the air supply was insufficient to maintain the required dissolved oxygen level, the inlet gas was switched from air to pure oxygen. Foaming was automatically suppressed by adding a 0.2% v/v aqueous emulsion of Sigma Antifoam SE-15 (Sigma-Aldrich, Burlington, MA, USA). Feeding was initiated after sufficient consumption of the initially supplied octanoic acid, operationally identified when the optical density reached approximately 10.0. Octanoic acid feeding began at the minimum pump flow rate of 1.67 mL/h, corresponding to approximately 10.5 mM/h, and was subsequently adjusted to prevent its accumulation above 10 mM. The feeding rate was determined using a carbon mass balance accounting for carbon utilization in residual biomass formation, PHO accumulation and CO₂ production [31,32]. Exhaust gas CO₂ was monitored as an indicator of culture activity and carbon availability. Ammonium sulfate was replenished based on frequent offline measurements to maintain its concentration within approximately 0.5–1.0 g/L. Phosphorus and magnesium availability were modified in the corresponding experiments to evaluate their effects on biomass formation, PHO accumulation and biopolymer properties. The bioreactor experiments were coded as B-CN, B-P and B-Mg, corresponding to continuous carbon and nitrogen feeding, phosphorus-limitation experiments and magnesium-feeding experiments, respectively.

2.3. Analytical Methods

Culture growth was monitored by measuring optical density (OD) at 600 nm using a Hitachi U-1800 (Tokyo, Japan) spectrophotometer. Dry cell weight (DCW) was determined gravimetrically from 10 mL culture samples after lyophilisation for 16 h. The intracellular PHO content of dried biomass samples was quantified using an FTIR-based method (PerkinElmer Inc., Waltham, MA, USA) previously described for the rapid screening and quantification of PHAs in bacterial biomass [41]. Glucose concentration was determined using the dinitrosalicylic acid (DNS) method [42]. Briefly, culture samples were centrifuged to remove the cells, and the resulting supernatants were appropriately diluted with distilled water. Each diluted sample was mixed with the DNS reagent and heated in a boiling water bath for 5 min to allow colour development. After cooling to room temperature, the absorbance was measured at 540 nm, and the glucose concentration was calculated using a calibration curve prepared with known glucose concentrations. Residual ammonium and phosphate concentrations were measured photometrically using Spectroquant Merck ammonium and phosphate test kits (Merck KGaA, Darmstadt, Germany), respectively. Magnesium concentration was implicitly inferred from net biomass formation using an experimentally derived mean biomass yield coefficient of 236 g DCW/g Mg reported in previously published studies on P. putida [31,36]. The composition of exhaust gases was measured online using a gas analyser equipped with a paramagnetic transducer for oxygen measurement and an infrared transducer for carbon dioxide measurement. These measurements were used to monitor the respiratory activity of the culture and to support the adjustment of the carbon feeding rate during fed-batch cultivation.

2.4. PHA Extraction and Purification

A combined mechanical–chemical extraction protocol was applied for cell disruption and recovery of the intracellularly accumulated PHO from P. putida cells, following the general principles of previously reported PHA extraction methods [43,44,45]. At the end of each culture, harvested cells were disrupted using a Sonics Vibra Cell VC-505 sonicator (Newtown, CT, USA) equipped with a titanium alloy probe. Sonication was performed in an ice bath using repeated sonication treatment–cooling cycles consisting of 30 s of sonication followed by 5 s of cooling. The total net sonication time was 30 min. After mechanical treatment, the disrupted cellular biomass was collected by centrifugation and the supernatant was discarded. Then, chloroform was added at a ratio of 10 mL of CHCl₃ per 50 mL of culture sample for cell resuspension and PHO extraction. The resulting suspension was gently stirred at 30 °C for 1.5 h and concentrated in a rotary evaporator. The dissolved PHO was then precipitated by adding cold methanol at a methanol-to-chloroform volume ratio of 10:1. Finally, for further purification, the precipitated biopolymer was redissolved in chloroform and centrifuged, after which the supernatant was poured onto a glass plate and allowed to dry in a desiccator.

2.5. PHA Characterization

The number (M_n)- and weight (M_w)-average molecular weights were determined using a gel permeation chromatograph (PolymerLabs GPC, Varian Inc., Palo Alto, CA, USA) equipped with an online viscometer and a refractive index detector. A universal calibration curve was constructed using polystyrene standards of known molecular weight. The extracted PHO samples were dissolved in a 95%:5% v/v mixture of chloroform and methanol and filtered through 1 μm Teflon filters before injection. The solvent flow rate and operating temperature were set at 1 mL/min and 35 °C, respectively. Thermal analysis of the produced biopolymers was performed using a differential scanning calorimeter (DSC-Q100, TA Instruments Ltd., New Castle, DE, USA). The glass transition (T_g) and melting (T_m) temperatures were determined from the recorded thermograms. Before DSC analysis, all samples were subjected to an initial scan to ensure a uniform thermal history. Thermal measurements were performed at a heating/cooling rate of 10 °C/min. Each sample was initially cooled to −75 °C and then heated to 70 °C. After cooling again to −75 °C, the sample was reheated to 70 °C at the same rate.

2.6. Statistical Analysis

All experiments were performed using three independent biological replicates (n = 3), and the results are presented as mean values ± standard deviation (SD). Statistical analysis was performed using Minitab Statistical Software v. 17.0 (Minitab Inc., State College, PA, USA). For comparisons involving three or more experimental conditions, the data were analysed using one-way analysis of variance (ANOVA), followed by Tukey’s honestly significant difference test for pairwise comparisons. Comparisons between two experimental conditions were performed using a two-sided Welch’s t-test. Differences were considered statistically significant when p < 0.05. Fixed experimental inputs, such as the selected initial nutrient concentrations, cultivation time and total amounts supplied, were not subjected to statistical analysis. Time-course data are presented descriptively as mean values ± SD, whereas inferential statistical comparisons were performed using the final cultivation measurements.

3. Results and Discussion

3.1. Effect of Carbon Source on Culture Growth, PHO Accumulation and Biopolymer Properties

The nature of the carbon source used is one of the most important factors affecting mcl-PHA production, as it governs not only biomass formation and intracellular accumulation but also the molecular structure and physicochemical properties of the recovered biopolymer [21,22,23,24,25,30,32]. In particular, structurally related carbon sources such as medium-chain fatty acids are metabolized through the β-oxidation pathway and serve as direct precursors for mcl-PHA biosynthesis. In contrast, structurally unrelated substrates, such as glucose, can support biomass formation and supply precursors for mcl-PHA biosynthesis through the de novo fatty acid synthesis pathway [25,30,37,46,47]. For this reason, octanoic acid, glucose, and their combination were initially evaluated in this study as carbon sources to examine their effect on culture growth, PHO accumulation and biopolymer molecular weight (MW).

As can be seen in

Table 1. Effect of carbon source on biomass growth, PHO accumulation, nutrient concentrations and molecular weight characteristics of the biopolymer produced by Pseudomonas putida KT2442. Values presented as mean ± SD of n = 3 independent biological replicates. Within each column, values that do not share a superscript letter are significantly different (p < 0.05).

Experiment Code: Carbon Source	Biomass (g/L)	PHO Content (% w/w)	Initial|Final Carbon Source Concentrations (g/L)	(NH4)2SO4 Final Concentration (g/L)	Mw (g/mol)	Mn (g/mol)
F-C1: Glucose	1.73 ± 0.09 b	10.02 ± 0.60 c	20.0|11.9 ± 0.6 b	0.002 ± 0.001 b	159,200 ± 8000 a	96,600 ± 4800 b
F-C2: Octanoic acid	1.82 ± 0.10 b	46.92 ± 1.90 b	3.6|0 a	0.091 ± 0.009 a	152,100 ± 7600 a	106,500 ± 5300 ab
F-C3: Glucose + octanoic acid	2.68 ± 0.13 a	56.15 ± 2.20 a	20.0|14.1 ± 0.8 b 3.6|0 a	0.007 ± 0.003 b	157,300 ± 7900 a	112,600 ± 5600 a

, when glucose was used as the sole carbon source, the final PHO content in biomass remained low, reaching approximately 10% w/w of the DCW, while the final biomass concentration was only 1.73 g/L after approximately 28 h of cultivation. Moreover, a considerable fraction of the initially supplied glucose remained unconsumed when the culture entered the stationary phase. This indicates that, under the tested conditions, glucose mainly supported residual biomass formation and was not an efficient standalone substrate for mcl-PHA accumulation by P. putida KT2442. Similar observations have been reported in other production systems where unrelated carbon sources promoted cell growth but resulted in relatively low mcl-PHA accumulation unless additional metabolic or feeding strategies are applied [29,35,46,47].

In contrast, when octanoic acid was used as the sole carbon source, PHO accumulation was strongly promoted (46.92% w/w), even though the final biomass concentration (1.82 g/L) remained relatively close to that obtained with glucose (Table 1). This confirms the role of octanoic acid as a structurally related precursor that is efficiently converted into C8-rich mcl-PHA through the β-oxidation pathway. The use of alkanoic acids has been repeatedly shown to influence mcl-PHA biosynthesis in P. putida, with substrate chain length affecting growth behaviour, bio-co-polymer composition and distribution, and biopolymer molecular properties [30,32]. FTIR/GC analysis identified 3-hydroxyoctanoate (C8) as the only PHA monomer in the recovered samples, while no additional PHA monomers were detected. Together with the molecular and thermal characterization results, these findings supported the identification of the recovered biopolymer as PHO. This is consistent with the well-established ability of Pseudomonas putida to convert octanoic acid into C8-rich mcl-PHA through the β-oxidation pathway [30,35,44].

The combined use of glucose and octanoic acid resulted in higher biomass formation (2.68 g/L) and higher PHO content (56.15% w/w) compared with the use of either substrate alone. This suggests that glucose primarily contributed to residual biomass formation, while octanoic acid acted as the main precursor for PHO synthesis. From a bioprocess engineering perspective, this observation highlights the advantage of partially decoupling biomass growth from biopolymer synthesis through the combined use of growth-supporting and biopolymer-directing carbon substrates. Similar conclusions have been drawn in recent studies, where co-feeding strategies using sugars or low-cost substrates together with fatty acids were proposed to enhance biomass growth, reduce production costs and control biopolymer properties [29,32,35]. Consequently, the present results demonstrate that combining structurally related and unrelated carbon sources represents a promising strategy for increasing PHO productivity while providing opportunities to tailor mcl-PHA characteristics for specific applications.

Statistical analysis confirmed that the carbon source significantly affected both biomass concentration and intracellular PHO content (p < 0.001). F-C3 resulted in significantly higher biomass than F-C1 and F-C2 and significantly higher PHO content than under both individual substrate conditions. Furthermore, F-C2 produced significantly more PHO than F-C1. No significant difference was observed in M_w (p = 0.551). However, M_n was significantly affected by the carbon source condition (p = 0.026), with F-C3 producing a higher value than F-C1. Thus, the carbon source primarily affected biomass formation and PHO accumulation, although a more limited effect on biopolymer chain length was also detected [15,16,17,18,19,20,25,32].

3.2. Effect of Initial Octanoic Acid Concentration

To further investigate the role of octanoic acid in PHO production, three different initial octanoic acid concentrations were tested in flask-scale experiments, namely, 15, 25 and 30 mM. Throughout this investigation, glucose was maintained at a fixed concentration of 20 g/L as a growth-supporting carbon source. This concentration was selected based on preliminary medium pre-optimization experiments, which indicated that 20 g/L provided reproducible, non-limiting glucose availability under the examined cultivation conditions, and was also consistent with concentrations previously applied in co-feeding and fed-batch studies with P. putida [31,32,33,34,40]. Glucose concentration was not examined as an independent optimization variable in this experimental series; rather, it was kept constant to isolate the effect of octanoic acid concentration and to maintain medium comparability with the subsequent bioreactor experiments.

As summarized in

Table 2. Effect of initial octanoic acid concentration on biomass growth, PHO accumulation and molecular weight of the biopolymer produced by Pseudomonas putida KT2442. Values presented as mean ± SD of n = 3 independent biological replicates. Within each column, mean values bearing different superscript letters are significantly different (p < 0.05).

Experiment Code: Octanoic Acid Concentration (mM)	Biomass (g/L)	PHO Content (% w/w)	Mw (g/mol)	Mn (g/mol)
F-OA1: 15	1.27 ± 0.06 b	55.0 ± 2.28 a	153,600 ± 7700 a	96,500 ± 4800 b
F-OA2: 25	2.68 ± 0.13 a	56.15 ± 2.20 a	157,300 ± 7900 a	112,600 ± 5600 a
F-OA3: 30	-	-	-	-

, increasing the initial octanoic acid concentration from 15 to 25 mM significantly increased biomass formation from 1.27 g/L to 2.68 g/L (p < 0.001), whereas intracellular PHO content remained statistically unchanged at approximately 55–56% w/w of DCW (p = 0.564). This suggests that, within this concentration range, additional carbon availability mainly supported culture growth, while the cells maintained a similar biopolymer accumulation capacity. Such behaviour is consistent with the growth-associated nature of mcl-PHA production in several Pseudomonas systems, where biopolymer accumulation can occur simultaneously with biomass formation when a suitable carbon source is available [28,36,37,38,39]. However, increasing the initial octanoic acid concentration to 30 mM completely prevented culture growth, demonstrating the toxic effects of excessive fatty acid concentration. In general, fatty acid toxicity represents a well-recognized limitation in mcl-PHA production, as the same substrates that serve as efficient biopolymer precursors may disrupt membrane integrity, impair their metabolism and inhibit microbial growth when present above critical concentrations [30,36,37,38,39,48]. This dual role explains why fed-batch strategies are commonly preferred for mcl-PHA production from fatty acids. Rather than adding a high initial substrate concentration, the carbon source should be supplied gradually to maintain a sufficient but non-inhibitory concentration range [36,37,38,39,48]. The increase in initial octanoic acid concentration did not significantly affect M_w (p = 0.592), whereas M_n was significantly higher in F-OA2 than in F-OA1 (p = 0.020). F-OA3 was excluded from the statistical comparisons because biomass formation and PHO production were not detected under this condition.

Overall, the results demonstrate that octanoic acid is an effective precursor for PHO accumulation by P. putida, but its concentration must be carefully controlled to maximize productivity without compromising culture viability. This finding is aligned with recent studies where the choice and concentration of alkanoic acids were shown to affect not only biopolymer accumulation but also monomer composition and MW characteristics [30,32,48]. Consequently, the carbon feeding strategy should be regarded not only as a productivity control tool but also as a potential product quality control variable.

3.3. Effect of Nitrogen Availability and Feeding

The effect of nitrogen availability on PHO production was investigated in flask-scale cultivations of P. putida KT2442 using the standard chemically defined medium supplemented with three different initial ammonium sulfate concentrations, namely 0.5, 1.0 and 1.5 g/L. As summarized in

Table 3. Effect of initial ammonium sulfate concentration and nitrogen pulse addition on biomass growth, PHO accumulation, residual nitrogen concentration and molecular weight characteristics of the biopolymer produced by Pseudomonas putida KT2442. Values presented as mean ± SD of n = 3 independent biological replicates. Within each column, values that do not share a superscript letter are significantly different (p < 0.05).

Experiment Code: (NH4)2SO4 Initial Concentration (g/L)	Biomass (g/L)	PHO Content (% w/w)	(NH4)2SO4 Final Concentration (g/L)	Mw (g/mol)	Mn (g/mol)
F-N1: 0.5	2.68 ± 0.13b	56.15 ± 2.30 a	0.007 ± 0.006 c	150,200 ± 7500 a	106,500 ± 5300 a
F-N2: 0.5 + 0.5	2.83 ± 0.14 b	51.31 ± 2.10 a	0.473 ± 0.047 b	154,600 ± 7700 a	114,400 ± 5700 a
F-N3: 1.0	3.39 ± 0.17 a	50.71 ± 3.00 a	0.434 ± 0.053 b	154,600 ± 6900 a	109,100 ± 6500 a
F-N4: 1.5	3.01 ± 0.13 b	50.07 ± 2.00 a	0.703 ± 0.081 a	146,200 ± 8300 a	90,500 ± 4900 b

, varying the initial nitrogen concentration within the tested range had no statistically significant effect on intracellular PHO accumulation (p = 0.0501). This suggests that, for this strain and under the specific conditions tested, nitrogen limitation was not essential for inducing substantial PHO accumulation. This observation contrasts with the conventional production strategy employed for many scl-PHA-producing bacteria, where biopolymer synthesis typically relies on nitrogen or phosphorus limitation following an initial biomass growth phase [7,8,9,10,40]. In contrast, mcl-PHA production by Pseudomonas species has frequently been associated with active cell growth and continuous carbon flux through fatty acid metabolism [28,36,37,38,39].

To further investigate the influence of nitrogen availability, an additional feeding strategy (F-N2) was evaluated, in which a single pulse replenishment of the culture with an additional 0.5 g/L of ammonium sulfate after 10 h of cultivation was applied. This experiment was designed to assess whether pulse nitrogen supplementation could enhance or adversely affect PHO production. Notably, the nitrogen pulse did not result in a meaningful improvement compared with experiment F-N3. Instead, the final biomass concentration was lower (2.83 versus 3.39 g/L), whereas the PHO content increased only marginally (51.31% versus 50.71% w/w). These results suggest that P. putida KT2442 is sensitive not only to overall nitrogen availability but also to the temporal profile of nutrient supply. Therefore, a medium may appear suitable based on its total nutrient composition, but the dynamic concentration profile experienced by the cells can redirect metabolism toward growth, storage, inhibition or stress response [36,37,38,39,49].

Increasing the initial ammonium sulfate concentration from 0.5 to 1.0 g/L significantly increased the final biomass concentration from 2.68 to 3.39 g/L. However, a further increase to 1.5 g/L did not provide an additional biomass benefit. Overall, nitrogen availability significantly affected biomass formation (p = 0.002), with F-N3 producing significantly more biomass than F-N1, F-N2 and F-N4. Although PHO content decreased numerically from 56.15% to approximately 50–51% w/w under the other nitrogen conditions, the overall difference did not meet the predefined threshold for statistical significance (p = 0.0501). Therefore, nitrogen availability clearly affected biomass formation, whereas its effect on intracellular PHO accumulation was statistically inconclusive. M_w remained unaffected (p = 0.507), while M_n was significantly affected (p = 0.004), primarily because F-N4 produced a lower M_n than the other nitrogen conditions. The significant differences in residual ammonium concentration (p < 0.001) further confirm that the cultures experienced different nitrogen-availability profiles [28,36,37,38,39,49].

3.4. Fed-Batch Cultivation with Continuous Carbon and Nitrogen Feeding

Following the flask-scale experiments, the cultivation was transferred to a benchtop fed-batch bioreactor. The aim was not only to increase biomass concentration and PHO production but also to evaluate whether controlled nutrient feeding could overcome the inhibitory effects associated with pulse substrate addition. Based on the flask-scale findings, both carbon and nitrogen were supplied continuously throughout the cultivation. Ammonium sulfate feeding was initiated at 0.3 g/(L·h) and subsequently adjusted according to the nitrogen requirements of the culture. Likewise, octanoic acid was initially supplied at the minimum pump flow rate (1.5 g/(L·h), corresponding to approximately 10.5 mM/h), and the feeding rate was continuously modified based on carbon mass balance calculations and online monitoring of CO₂ evolution, thereby maintaining the fatty acid concentration below the inhibitory threshold.

Under the above strategy, the process reached a final DCW of approximately 20 g/L after 27 h of cultivation, while PHO content remained close to 50% w/w (Figure 1). Compared with the flask-scale experiments, this represents a clear improvement in performance, mainly due to enhanced biomass formation under controlled conditions. Importantly, the increased biomass concentration was achieved without compromising biopolymer accumulation, confirming that higher PHO production can be obtained by sustaining active growth and avoiding carbon source inhibition rather than by imposing severe nutrient limitation [28,36,37,38,39]. The nutrient concentration profiles further demonstrate the effectiveness of the feeding strategy (Figure 2). Under the applied fed-batch strategy, nitrogen concentration remained close to the desired range until the late stages of the culture, when nitrogen accumulation indicated that growth had slowed or stopped. Similarly, octanoic acid concentration remained relatively stable and did not exhibit sharp increases, confirming the effectiveness of the continuous feeding strategy. Although phosphorus concentration gradually declined during cultivation, it was not depleted, indicating that phosphorus did not become growth-limiting under the applied conditions.

Despite the dynamic changes in nutrient concentrations, intracellular PHO content remained relatively constant, suggesting that biopolymer synthesis was sustained as long as the microbial cells were maintained in a favourable metabolic state. These observations reinforce the conclusions drawn from the flask-scale experiments, namely that stable PHO production depends primarily on maintaining balanced nutrient availability and controlled substrate delivery rather than on inducing severe nutrient limitation. Moreover, as is evident in Figure 3, dissolved oxygen (D.O.) was maintained at 25% of air saturation via agitation control and, when necessary, oxygen supplementation. The need for pure oxygen indicates that oxygen transfer became a relevant constraint at higher biomass concentrations. At the same time, the evolution of CO₂ concentration in the exhaust gas reflected the respiratory activity of the culture and served as a reliable indicator for adjusting octanoic acid feeding. This type of bioprocess monitoring is highly relevant for future scale-up, because mcl-PHA production from fatty acids requires simultaneous control of carbon availability, oxygen transfer capacity and biomass growth [36,37,38,39,50].

Recent fed-batch studies confirm that feeding strategy is a decisive element in mcl-PHA production. Approaches such as D.O.-stat control or substrate-limited fed-batch operation have been used to improve bioprocess stability and productivity in P. putida cultures grown on crude glycerol or substrates based on fatty acids [28,33,36,37,38,39,49,50]. The broader conclusion from these studies is that the principal advantage of fed-batch cultivation is not merely the achievement of higher biomass concentrations but the ability to maintain cells in a productive physiological state through dynamic regulation of nutrient availability. The present work is consistent with this concept and demonstrates that a simple but carefully adjusted continuous feeding policy effectively prevents substrate inhibition, sustains biomass growth and markedly enhances PHO production compared with the pulse-feeding approach evaluated at flask scale.

3.5. Effect of Phosphorus Limitation

Following the establishment of a stable fed-batch cultivation strategy based on continuous carbon and nitrogen feeding, phosphorus availability was investigated as a potential means of enhancing PHO production and modifying biopolymer synthesis. To this end, the initial phosphate salt concentration was reduced from the standard level (11 g/L in total) to 2.2 g/L and 1.1 g/L, while the continuous feeding approach for octanoic acid and ammonium sulfate remained unchanged. Figure 4 and Figure 5 present the dynamic profiles and performance metrics of the cultivation carried out with reduced phosphate concentration (2.2 g/L), while Figure 6 and Figure 7 present the corresponding profiles for the cultivation performed under stronger phosphorus limitation (1.1 g/L). In both phosphorus-limited cultivations, biomass increased during the initial cultivation phase and PHO accumulation followed the growth profile. Carbon and nitrogen remained available throughout the cultivations, whereas phosphate was depleted before the end of the process. Following phosphate depletion, biomass formation ceased rapidly despite the continued availability of the other nutrients, indicating that phosphorus became the primary growth-limiting substrate under these operating conditions.

As can also be seen in

Table 4. Comparison of the final cultivation performance, PHO production and biopolymer properties of Pseudomonas putida KT2442 cultivated in a fed-batch bioreactor under different phosphorus availability conditions. Values presented as mean ± SD of n = 3 independent biological replicates. Within each row, values that do not share a superscript letter are significantly different (p < 0.05).

Parameter	B-CN	B-P1	B-P2
Initial phosphate salt concentration (g/L)	11.0	2.2	1.1
Cultivation time (h)	27.0	27.0	30
Total octanoic acid supplied (mM)	265	225	144
Octanoic acid consumed (mM)	177 ± 7.1 a	145 ± 7.9 b	125 ± 5.5 c
Total (NH4)2SO4 supplied (g/L)	10.05	8.90	6.50
(NH4)2SO4 consumed (g/L)	8.00 ± 0.40 a	6.20 ± 0.32 b	4.66 ± 0.25 c
Phosphate consumed (g/L)	6.0 ± 0.5 a	2.2 ± 0.2 b	1.1 ± 0.1 c
Final dry cell weight (DCW) (g/L)	20.39 ± 1.11 a	16.85 ± 1.21 b	14.38 ± 0.87 b
PHO content (% w/w DCW)	50 ± 2.5 ab	45 ± 3.0 b	53 ± 3.8 a
PHO concentration (g/L)	10.19 ± 0.70 a	7.58 ± 0.70 b	7.62 ± 0.61 b
PHO productivity (g/(L·h))	0.377 ± 0.036 a	0.281 ± 0.024 ab	0.254 ± 0.021 b
Weight-average molecular weight, Mw (g/mol)	131,200 ± 6600 a	141,300 ± 7100 a	129,400 ± 6500 a
Number-average molecular weight, Mn (g/mol)	84,900 ± 4200 a	80,700 ± 3900 a	82,900 ± 4100 a
Polydispersity index (PDI)	1.54 ± 0.08 b	1.75 ± 0.09 a	1.56 ± 0.07 ab
Glass transition temperature, Tg (°C)	−36.7 ± 0.3 b	−37.22 ± 0.3 b	−35.3 ± 0.3 a
Melting temperature, Tm (°C)	53.7 ± 0.4 a	52.8 ± 0.5 ab	52.4 ± 0.4 b

, the reduction in phosphorus availability had a statistically significant effect on biomass formation. Decreasing the initial phosphate salt concentration from 11.0 to 2.2 and 1.1 g/L reduced final DCW from 20.39 to 16.85 and 14.38 g/L, respectively (p = 0.001). Both phosphorus-limited conditions produced significantly less biomass than B-CN, whereas B-P1 and B-P2 did not differ significantly from each other. PHO concentration was also significantly reduced under both phosphorus-limited conditions (p = 0.005). The overall effect on PHO content did not reach statistical significance (p = 0.054), and neither phosphorus-limited condition differed significantly from B-CN. PHO productivity was significantly affected by phosphorus availability (p = 0.032), with B-P2 producing a lower value than B-CN. No significant differences were detected in M_w (p = 0.144) or M_n (p = 0.492). Nevertheless, significant but relatively small differences were observed in PDI (p = 0.034), T_g (p < 0.001) and T_m (p = 0.027).

The above findings suggest that, under the conditions examined, phosphorus availability primarily regulated biomass formation rather than carbon partitioning towards biopolymer synthesis. This is important because it challenges the assumption that nutrient limitation necessarily enhances PHA and specifically PHO accumulation [7,8,9,10,40]. This behaviour is consistent with the idea that PHO production in Pseudomonas can remain growth-associated and that excessive nutrient restriction may reduce the number of active biopolymer-producing cells rather than redirecting metabolism toward higher accumulation [28,36,37,38,39]. The phosphorus consumption profiles also showed that phosphate uptake was significantly influenced by the initial phosphate availability (Figure 8). For approximately the same level of biomass production and biopolymer accumulation, different amounts of phosphate were consumed depending on the initial phosphate concentration. This suggests that phosphorus uptake was not only linked to biomass formation but also influenced by the extracellular nutrient environment. This trend has practical relevance for cultivation medium design, because excessive phosphate may not be necessary, but insufficient phosphate can prematurely terminate growth and reduce PHA production.

Overall, the phosphorus limitation experiments indicate that phosphorus should be controlled as a growth-supporting nutrient rather than used as a strong limitation trigger in this bioprocess. From a material perspective, this is also relevant because reducing biomass formation without improving biopolymer content or molecular properties does not support the production of a more valuable biopolymer grade. Together with the carbon and nitrogen feeding experiments, the present results further reinforce the central conclusion of this work that balanced nutrient availability and controlled substrate delivery are more effective for maximizing PHO production than abrupt or severe nutrient limitation.

3.6. Effect of Magnesium on Fed-Batch PHO Production

Since neither nitrogen nor phosphorus limitation enhanced PHO accumulation, magnesium availability in the culture was examined as an alternative process variable to support biomass formation and potentially influence biopolymer production and molecular properties. Generally, magnesium plays a central role in microbial physiology, acting as a cofactor in numerous enzymatic reactions and contributing to ribosome stability, energy metabolism and cellular stress responses. Its importance may become particularly relevant under high-cell-density cultivation conditions, where cellular demand for essential minerals increases [51]. In PHA-producing systems, magnesium has also been associated with intracellular biopolymer granule formation and depolymerase activity, although its role in mcl-PHA production remains less well understood compared with carbon and nitrogen availability [23,52,53].

It should be noted that magnesium was initially supplied as a single pulse addition of MgSO₄ during the exponential growth phase. However, this strategy negatively affected the culture, resulting in temporary growth suspension and reduced PHO content. This response is consistent with the broader behaviour observed for abrupt carbon and nitrogen supplementation, indicating that sudden nutrient concentration shifts can disturb the physiological balance of P. putida KT2442 [36,37,38,39,49].

Therefore, continuous magnesium supplementation was subsequently implemented together with the already-established octanoic acid and ammonium sulfate feeding strategy. As shown in Figure 9, continuous magnesium feeding improved overall process performance without any negative impact on culture growth. Specifically, under this strategy, the final biomass concentration increased from 20.4 to 26 g/L within approximately 30 h of cultivation, while PHO content remained close to 50% w/w during most of the cultivation. A slight reduction in PHO content was observed only during the final stage of cultivation, suggesting a shift in the balance between biomass formation and biopolymer accumulation as magnesium availability increased. Continuous magnesium feeding significantly increased final DCW from 20.39 to 26.00 g/L (p = 0.006). The final PHO concentration increased numerically from 10.19 to 11.50 g/L, but this difference was not statistically significant (p = 0.131). Similarly, PHO productivity remained statistically unchanged (p = 0.809). The cultivation profiles shown in Figure 10 further indicate that magnesium availability influenced the relationship between cell growth and PHO accumulation during the late cultivation phase. Although the exact mechanism cannot be elucidated from the present data alone, the results suggest that magnesium should not be treated only as a fixed medium component. Instead, its feeding profile may represent a process control parameter affecting biomass–PHO partitioning and product quality [52,53].

The effect of continuous magnesium supplementation is also summarized in

Table 5. Comparison of the final cultivation performance, PHO production, and biopolymer properties of Pseudomonas putida KT2442 cultivated in a fed-batch bioreactor with and without continuous magnesium supplementation. Values presented as mean ± SD of n = 3 independent biological replicates. Within each row, mean values bearing different superscript letters are significantly different (p < 0.05).

Parameter	B-CN	B-Mg
Cultivation time (h)	27.0	30.0
Total octanoic acid supplied (mM)	265	320
Octanoic acid consumed (mM)	177 ± 7.1 b	250 ± 10.3 a
Total (NH4)2SO4 supplied (g/L)	10.05	12.80
(NH4)2SO4 consumed (g/L)	8.00 ± 0.40 b	11.07 ± 0.54 a
Total magnesium supplied (g/L)	6.0	1.1
Magnesium consumed (g/L)	0.8 ± 0.04 b	0.2 ± 0.06 a
Final dry cell weight (DCW) (g/L)	20.39 ± 1.11 b	26.00 ± 1.38 a
PHO content (% w/w DCW)	50 ± 2.5 a	44 ± 3.0 a
PHO concentration (g/L)	10.19 ± 0.70 a	11.50 ± 0.94 a
PHO productivity (g/(L·h))	0.377 ± 0.036 a	0.384 ± 0.030 a
Weight-average molecular weight, Mw (g/mol)	131,200 ± 6600 a	145,000 ± 7300 a
Number-average molecular weight, Mn (g/mol)	84,900 ± 4200 a	82,000 ± 4100 a
Polydispersity index (PDI)	1.54 ± 0.08 b	1.74 ± 0.09 a
Glass transition temperature, Tg (°C)	−36.7 ± 0.3 a	−36.06 ± 0.3 a
Melting temperature, Tm (°C)	53.7 ± 0.4 a	52.0 ± 0.6 b

. Magnesium feeding significantly increased biomass formation (p = 0.006), whereas PHO content (p = 0.058), PHO concentration (p = 0.131) and PHO productivity (p = 0.809) were not significantly affected. Thus, magnesium promoted biomass formation while maintaining PHO accumulation, but a statistically significant improvement in volumetric PHO production was not demonstrated. Neither M_w (p = 0.073) nor M_n (p = 0.440) differed significantly between B-CN and B-Mg. PDI increased significantly (p = 0.046), while T_g remained unchanged (p = 0.059). A small but statistically significant reduction was detected in T_m (p = 0.020). Overall, continuous magnesium addition primarily influenced biomass formation rather than PHO accumulation or volumetric productivity. The significant increase in biomass, together with the absence of significant differences in PHO content, concentration and productivity, suggests that magnesium improved culture growth but did not significantly enhance PHO production under the examined conditions. Nevertheless, magnesium-feeding dynamics remain relevant as an additional process variable for future optimization. Compared with recent fed-batch mcl-PHA studies, the main value of the present magnesium feeding experiment is not only the significant increase in biomass and the numerical increase in final PHO concentration, but also the demonstration that mineral feeding dynamics can substantially influence bioprocess performance [28,33,49,50]. While most of these studies focus on carbon source selection, strain engineering or nitrogen-limited production stages, the present results show that magnesium addition can also be used as a bioprocess variable. This provides an additional degree of freedom for future optimization of mcl-PHA production and biopolymer grade control.

3.7. Effect of Nutrient Feeding on Molecular Weight and Thermal Properties

The molecular characteristics of PHO are important quality parameters, as they influence biopolymer processability, mechanical performance, degradation kinetics and application suitability [13,14,15,16,17,18,19,20,21,22,23,24,25,54,55]. Since molecular weight can be affected by multiple factors, including the producing strain, substrate nature and availability, nutrient balance, PHA synthase activity, intracellular depolymerase activity and cultivation conditions [9,15,16,17,18,19,20,25,52,53,54,55], the present study evaluated not only PHO production but also molecular and thermal properties of the recovered biopolymer. Selected samples from the bioreactor experiments were therefore analysed to assess the potential influence of octanoic acid availability and magnesium supplementation on biopolymer characteristics.

The obtained results indicate that the molecular weight characteristics exhibited some variation among the different feeding conditions (Figure 11 and Figure 12). In particular, changes in octanoic acid availability during cultivation were accompanied by variations in the measured molecular weight values, suggesting that substrate feeding conditions may influence biopolymer chain formation and/or its intracellular turnover. However, the observed differences do not demonstrate a clear linear relationship between octanoic acid concentration and molecular weight, indicating that additional physiological factors likely contribute to the final biopolymer characteristics. Similar observations have been reported for mcl-PHA-producing Pseudomonas strains, where molecular weight distribution is influenced by the combined effects of substrate supply, metabolic state, polymerization activity, and intracellular degradation processes [15,16,17,18,19,20,52,53,54,55].

A comparable trend was observed when octanoic acid was continuously co-fed with magnesium. Under these conditions, the biopolymer molecular weight profile appeared somewhat more stable compared with the cultivation without magnesium addition (Figure 12). Although the present data do not allow a direct attribution of this effect to magnesium availability, the results suggest that mineral supplementation may contribute to maintaining a more consistent cellular environment during high-density cultivation. Thus, the nutrient feeding strategy may represent one of several factors affecting biopolymer molecular characteristics, although further systematic studies would be required to establish specific relationships between individual nutrients and molecular weight control.

Across the examined experiments, the produced PHO showed number-average molecular weight (M_n) values in the range of approximately 70,000–130,000 g/mol and weight-average molecular weight (M_w) values between approximately 115,000 and 280,000 g/mol, with polydispersity index values ranging from 1.3 to 2.3. These values include final biopolymer samples and samples collected at intermediate cultivation stages, reflecting the dynamic evolution of biopolymer characteristics throughout the cultivation process.

The thermal properties of the selected PHO samples were also evaluated. In all cases, the glass transition temperature (T_g) and melting temperature (T_m) were around −36 °C and 52 °C, respectively. These values are consistent with the expected behaviour of mcl-PHAs, which typically show low glass transition temperatures and elastomeric characteristics due to their medium-chain-length monomer composition [13,21,22,23,24,25,55,56,57]. Furthermore, the absence of a distinct melting peak during the second heating scan indicates that the produced material exhibited predominantly amorphous or weakly crystalline characteristics under the applied thermal history. Similar thermal characteristics have been reported for mcl-PHAs produced by Pseudomonas species using fatty acid or other renewable substrates [25,55,56,57]. Overall, the absolute variations in the thermal properties of the produced PHO were relatively small. Nevertheless, statistically significant differences in T_g and/or T_m were detected under selected phosphorus and magnesium conditions. Similarly, M_w was generally stable, but significant differences in M_n and PDI were detected in selected comparisons. Therefore, the cultivation conditions generally preserved the characteristic molecular and thermal profile of PHO, although the measured properties were not statistically identical under all conditions.

From a commercial perspective, further improvement of PHO production will require the simultaneous intensification of upstream cultivation and downstream recovery. High-cell-density, substrate-limited fed-batch operation combined with automated feeding control can minimize fatty acid inhibition and maintain the cells in a productive physiological state [31,32]. Co-feeding lower-cost substrates with a carefully controlled octanoic acid precursor stream may also reduce raw material requirements while sustaining biomass formation and PHO accumulation [33,34]. At larger scales, oxygen transfer capacity and dissolved oxygen control must be integrated with balanced nitrogen, phosphorus and magnesium availability to prevent premature growth arrest or abrupt nutrient-induced stress. Commercial feasibility will also depend on efficient cell disruption and PHO recovery, reduced solvent consumption, solvent recycling and the implementation of simplified or alternative extraction methods [4,5,6,23,44,45]. Finally, maintaining reproducible molecular weight distributions and thermal properties will be essential for producing application-specific PHO grades. The controlled feeding strategies examined in this study provide a basis for this development, although pilot-scale validation and techno-economic assessment will be required.

4. Conclusions

This study systematically evaluated the effects of carbon, nitrogen, phosphorus and magnesium feeding strategies on biomass formation, PHO production and biopolymer properties during the cultivation of Pseudomonas putida KT2442. The results demonstrate that bioprocess performance is governed primarily by nutrient availability and feeding dynamics rather than by severe nutrient limitation. Octanoic acid proved to be an effective precursor for C8 mcl-PHA biosynthesis, provided that its concentration was maintained below inhibitory levels through controlled feeding. Similarly, sharp changes in nutrient concentrations, especially during pulse additions of carbon, nitrogen or magnesium sources, adversely affected culture performance, highlighting the importance of stable nutrient supply throughout the cultivation.

Under the examined conditions, neither nitrogen nor phosphorus limitation significantly enhanced intracellular PHO accumulation. Severe phosphorus limitation significantly reduced biomass formation and final PHO concentration. In contrast, continuous carbon and nitrogen feeding supported high-cell-density cultivation and PHO production. Continuous magnesium supplementation significantly increased biomass concentration while maintaining intracellular PHO content; however, the differences in final PHO concentration and volumetric productivity were not statistically significant. Magnesium supplementation was beneficial only when applied continuously, whereas pulse addition adversely affected culture performance.

The feeding strategies generally produced PHO within a comparable molecular and thermal property range. Nevertheless, selected statistically significant differences were detected in molecular and thermal properties, depending on the investigated nutrient condition. These differences were relatively small in absolute terms but demonstrate that the recovered PHO was not statistically identical under all cultivation conditions. Further investigation using larger datasets, together with mechanical and rheological characterization, will be required to establish quantitative relationships between nutrient-feeding profiles and application-relevant biopolymer quality.

Overall, the present work demonstrates that controlled fed-batch nutrient delivery provides a robust strategy for improving PHO production by P. putida KT2442 while generally preserving the characteristic molecular and thermal profile of PHO. These findings contribute to the development of more efficient mcl-PHA production processes and provide a basis for future optimization through advanced feeding strategies, process modelling, and evaluation of the relationship between biopolymer molecular characteristics and end-use material performance.