Production of Potential Substitutes for Conventional Plastics Using Metabolically Engineered Acetobacterium woodii

Abstract

Increasing greenhouse gas emissions and decreasing fossil fuel supplies necessitate the development of alternative methods for producing petroleum-based commodities. Plastics are also primarily petroleum-based goods with rising demand, thus there is growing interest in plastic substitutes. Polyhydroxyalkanoates (PHAs) are naturally produced biopolymers that are utilized by microorganisms as a source of energy and carbon storage. Poly-3-hydroxybutyrate (PHB) is a member of the PHA family and is considered the most promising candidate to replace polyethylene (PE). PHB is naturally produced by Cupriavidus necator, but recombinant production has also been recently established. This study is the first to investigate the heterologous production of PHB with recombinant Acetobacterium woodii using CO₂ + H₂ as a carbon and energy source. The introduction of a synthetic PHB production pathway resulted in the production of 1.23 g/L CDW and 1.9% PHB/cell dry weight (CDW), which corresponds to a production of 23.5 mg/L PHB. PHB quantification was simplified using LipidGreen2 fluorescence measurements.

1. Introduction

Plastics have become an integral part of modern society, contributing to various industries due to their versatility and cost-effectiveness [1]. Although the production of petroleum-based plastics has a severe effect on the climate crisis and plastic pollution, the demand and production of conventional plastics have increased dramatically. In the year 2021, worldwide plastics production amounted to 390.7 million metric tons (Mt), which implies an increase of 15.2 Mt in comparison to the production of 2020 [2]. In recent years, bioplastics have emerged as a potential alternative, offering the prospect of reducing dependence on fossil fuels and mitigating plastic pollution. The family of bioplastics comprises three types of plastics: plastics derived from renewable resources, biodegradable plastics, and plastics that are both bio-based and biodegradable [3]. Common types of bioplastics include polylactic acid (PLA), starch-based polymers, and polyhydroxyalkanoates (PHAs) [4]. PHAs are produced by numerous microorganisms as intracellular storage compounds in the form of granules and serve as carbon and energy reserves. These polymers are composed of repeating units of hydroxyalkanoic acids, which can vary in chain length and composition [5]. Hydroxyalkanoic acids are naturally synthesized from various carbon sources, such as glucose, fructose, or fatty acids, through the metabolic pathways of microorganisms. PHAs are considered an attractive alternative to conventional petrochemical-based plastics due to their biodegradability, renewable nature, and potential for sustainable production [4]. However, substrate costs and competition with human nutrition render the production of PHAs at an industrial scale not economically competitive to the production of conventional plastics. Single carbon gases (C1-gases) such as CO or CO₂ (+ H₂) provide a cheap alternative as substrate and can be metabolized by acetogens [6]. Acetogens comprise a huge variety of bacteria in many different genera, and all use the Wood–Ljungdahl pathway for CO₂ fixation [7]. The acetogenic bacterial species Clostridium coskatii, C. ljungdahlii, and C. autoethanogenum have already been employed to produce the PHA poly-3-hydroxybutyrate (PHB) by using synthesis gas (consisting of CO, CO_2, and H₂) as a carbon source [7,8]. For PHB production in C. autoethanogenum, the production pathway of the natural producer Cupriavidus necator was codon-optimized and expressed as plasmid-based, resulting in a total production of 5.6% PHB per cell dry weight (CDW) [9].

Another challenging problem of PHB production via bacterial fermentation is caused by time-consuming quantification methods and costly isolation procedures [10]. Common quantification methods include lyophilization of cell material, cell lysis, and afterward chromatographic analysis, which lasts at least several days [11,12]. Recently, a different quantification method for PHB production in C. necator was established by Kettner et al. using the fluorescent dye LipidGreen2 (LG2), which binds to neutral lipids only [10,13,14].

In this study, recombinant PHB production with acetogens using CO₂ + H₂ was investigated for the first time. Therefore, a synthetic pathway for PHB production was expressed in Acetobacterium woodii, using the pathway of C. acetireducens and replacing the thlA, hbd, and crt genes with those of C. scatologenes. During fermentation, PHB production was visualized and monitored via fluorescence measurements. Hence, a modified version of the LG2 assay for quantification of PHB in A. woodii was established.

2. Materials and Methods

2.1. Bacterial Strains and Media

Bacterial strains used in this work are listed in

Table 1. Bacterial strains, their genotype, and origin used in this study.

Strain	Genotype	Origin
E. coli XL1-Blue MRF’	Δ (mcrA)183 Δ (mcrCB-hsd SMR-mrr)173 endA1 supE44 thi-1recA1 gyrA96 relA1 lac [F’proAB lacIq ZΔM15 Tn10 (TetR)]	Agilent Technologies (Santa Clara, CA, USA)
A. woodii DSM 1030	type strain	German Collection of Microorganisms and Cell Cultures (DSMZ, Brunswick, Germany)

. Plasmid amplification was carried out in Escherichia coli XL1-Blue MRF’ strains. Cultivation of E. coli strains was performed in Lysogeny broth (LB) medium (containing per liter: tryptone 10 g, NaCl 10 g, and yeast extract 5 g) supplemented with the respective antibiotic aerobically at 37 °C with shaking at 180 rpm [15].

Acetobacterium woodii DSM 1030 strains were cultivated anaerobically at 30 °C in modified DSM medium 135 (containing per liter): NH₄Cl 0.2 g, KH₂PO₄ 1.76 g, K₂HPO₄ 8.44 g, yeast extract 3 g, NaHCO₃ 10 g, HCl-cysteine monohydrate 0.3 g, Na₂S nonahydrate 0.3 g, trace element solution 2 mL, vitamin solution 2 mL, and resazurin 1 mg; trace element solution (per liter): nitrilotriacetate 12,8 g, NaOH 0,5 g, MnCl₂ tetrahydrate 0.1 g, NaCl 5 g, FeCl₂ tetrahydrate 2 g, CoCl₂ hexahydrate 0.2 g, ZnCl₂ 70 mg, CuCl₂ dihydrate 2 mg, H₃BO₃ 6 mg, Na₂MoO₄ dihydrate 36 mg, NiCl₂ hexahydrate 24 mg, Na₂SeO₃ pentahydrate 3 mg, and Na₂WO₄ dihydrate 4 mg; vitamin solution (per liter): HCl-pyridoxine 50 mg, HCl-thiamine monohydrate 50 mg, riboflavin 50 mg, D-Ca-pantothenate 50 mg, lipoic acid 25 mg, p-aminobenzoate 50 mg, nicotinic acid 50 mg, vitamin B₁₂ 25 mg, biotin 25 mg, and folic acid 25 mg). After autoclaving at 121 °C and 1.2 bar for 15 min, 1.3 mM MgSO₄, 40 mM fructose, and the respective antibiotic were added [16].

2.2. Isolation of Genomic and Plasmid DNA

The genomic DNA of A. woodii strains was isolated using the MasterPure^TM GramPositive DNA Purification Kit (Epicentre, Madison, WI, USA). An amount of 2 mL of late exponential cultures was harvested via centrifugation (1 min, 13,000× g, 4 °C) and further processed as described by the manufacturer. Plasmid DNA was isolated from E. coli strains using the Zyppy^TM Plasmid Miniprep Kit (ZYMO Research Europe GmbH, Freiburg, Germany). An amount of 4 mL of an overnight culture was centrifuged (13,000× g, 1 min, RT) and further processed as described by the manufacturer’s instructions.

2.3. Electroporation Procedure for A. Woodii

Electrocompetent A. woodii cells were prepared and transformed as previously described [16]. For verification of transformed strains, genomic DNA was isolated, and 16 S rDNA was amplified via PCR and sent for sequencing. Plasmid integrity was verified via retransformation of E. coli XL1-Blue MRF’, plasmid DNA isolation, and restriction digestion.

2.4. Growth Conditions of Batch Experiments

Cultivation of A. woodii strains was performed anaerobically in a modified DSM 135 medium at 30 °C. Heterotrophic growth experiments were performed using a 200 mL medium filled into 500 mL serum bottles (SGD Pharma, Paris, France), and 60 mM fructose was added as a carbon and energy source. For autotrophic growth, the gas phase of 1000 mL serum bottles (SGD Pharma, Paris, France) filled with 200 mL medium was changed to 1 bar CO₂ + H₂ (33% + 67%, MTI Industriegase AG, Elchingen, Germany). The bottles were incubated with shaking at 130 rpm. The gas phase was refilled at a minimal pressure of 0.3 bar.

Before the growth experiments started, A. woodii strains were adapted to the respective cultivation conditions by transferring them once into a fresh medium. Before and after growth strains were verified as performed after the transformation of electrocompetent A. woodii cells.

Growth experiments were performed on biological triplicates. During growth, OD₆₀₀, pH, substrate consumption, and end-product concentrations were monitored. Furthermore, samples for fluorescence measurements and PHB analysis were taken.

For the establishment of a LipidGreen2 assay, recombinant A. woodii was cultivated in a 1.5 L modified DSM 135 medium filled into three 1000 mL serum bottles with 60 mM fructose added. Bottles were incubated with shaking at 130 rpm.

2.5. Analytics

2.5.1. Optical Density and pH Measurements

The optical density of cultures during growth experiments was determined using the Ultrospec 3100 pro UV/Visible Spectrophotometer (Amersham Biosciences Europe GmbH, Freiburg, Germany) at a wavelength of 600 nm. The change in pH of the culture broth was measured using the pH-meter pH 522 (WTW- a xylem brand, Weilheim, Germany). Before the determination of the pH, the pH meter was calibrated using two-point calibration at pH 7 and 4.

2.5.2. High-Performance Liquid Chromatography

During heterotrophic and autotrophic growth experiments, fructose consumption, and acetate production were monitored via HPLC analysis using the Agilent 1260 Infinity Series HPLC System (Agilent Technologies, Santa Clara, CA, USA). An amount of 500 µL of cell suspension was centrifuged at 13,000× g for 30 min at 4 °C and the supernatant was transferred to a glass vial (CS Chromatographie Service GmbH, Langerwehe, Germany) closed with a crimp cap (BGB Analytik Vertrieb, Rheinfelden, Germany). Analysis was performed using a diode array detector and a refractive index detector at a temperature of 35 °C in combination with the CS-Chromatographie organic acid column with a length of 150 mm (CS-Chromatographie Service GmbH, Langerwehe, Germany) operating at 40 °C. Separation of 20 µL injection volume occurred with 5 mM H₂SO₄ as mobile phase with a flow of 0.7 mL per min. Commercially available acetate and fructose served as external calibration standards.

2.5.3. Establishment of a LipidGreen2-Assay for In Vivo PHB Detection

Detection of intracellularly stored PHB via fluorescence measurements was performed using the fluorescent dye LipidGreen2 (Sigma Aldrich, Steinwehe, Germany). For fluorescence measurement, 2 mL samples were taken during heterotrophic and autotrophic growth experiments. The OD₆₀₀ of the cell suspension was normalized in phosphate-buffered saline (PBS, per liter: NaCl 8.0 g, KCl 0.2 g, Na₂HPO₄ 1.4 g, KH₂PO₄ 240 mg; pH was set to 7.4) to different values between 0.1 and 1 in a sample volume of 500 µL. A sample volume of 100 µL served as a blank value for fluorescence intensity and OD measurements using a black 96-well plate (Greiner Bio-One GmbH, Frickenhausen, Germany) and a transparent 96-well plate (Sarstedt AG & CoKG, Nümbrecht, Germany), respectively. The remaining cell suspension was stained for 10 min with LipidGreen2 (in DMSO) at room temperature and with light exclusion in different concentrations ranging from 10.8 µM to 108 µM. Afterward, fluorescence intensity was either measured or the cell suspension was washed once with PBS via centrifugation (5 min, 2500× g, 4 °C). Fluorescence intensity was measured using the BioTek Synergy H1 plate reader (Agilent Technologies, Santa Clara, CA, USA) at an excitation wavelength of 440 nm and an emission wavelength of 505 nm. For the comparability of fluorescence intensities, the gain was set to a value of 70. After the fluorescence measurement, the optical density was checked again at wavelengths of 600 nm and 660 nm.

2.5.4. Quantification of PHB Using Gas Chromatography

To assign the received fluorescence intensities to the affiliated PHB content/cell dry weight (PHB/CDW), PHB had to be quantified via gas chromatography. A modified protocol of Flüchter et al. was used [6]. During growth, 10–200 mL of culture broth was taken and cells were harvested via centrifugation for 10 min at 4 °C and 4000× g. Cell sediment was suspended in 0.8% saline and washed once by centrifugation for 10 min at 4 °C and 4000× g. The pellet was stored at −20 °C until further usage. For lyophilization, the cell pellet was suspended in 1–5 mL 0.8% saline. The suspension was transferred to heat-proof glass vials (Bellco Glass Inc., Vineland, NJ, USA) and closed with cotton cellulose rolls (HenrySchein Services GmbH, Langen, Germany). Cells were shock-frozen in liquid N₂ and placed into a −80 °C pre-cooled cooling-block. Lyophilization was performed for 24–72 h at −54 °C and 0.01 mbar using the lyophilizator Alpha 1–4 LDplus (Martin Christ Gefriertrocknungsanlagen GmbH, Osterode am Harz, Germany). After lyophilization, cotton cellulose rolls were replaced by screwcaps and biomass was stored at 4 °C until further processing. An amount of 10–50 mg of lyophilized cell mass was transferred to Hungate tubes (Bellco Glass inc., Vineland, NJ, USA), incubated with 2 mL of freshly prepared acidic methanol (85% methanol, 15% sulfuric acid), and mixed thoroughly. Afterward, 2 mL of anhydrous chloroform was added, and the suspension was mixed thoroughly again. The reaction mixture was incubated for 4 h at 100 °C in an oil bath and afterward left to cool at room temperature. An amount of 2 mL of water was added, and the suspension was mixed thoroughly. After phase separation, the organic phase was transferred to glass vials (CS-Chromatographie Service GmbH, Langerwehe, Germany) and sealed with screwcaps. Gas chromatography was performed using the gas chromatograph Perkin Elmer Clarus 680 GC system (Perkin Elmer, Waltham, MA, USA). The chromatograph was equipped with an Elite-FFAP capillary column (Perkin Elmer, Waltham, MA, USA) with a length of 30 m, inner diameter of 0.32 mm, and a density film of 0.25 µM. The injection volume was set to 1 µL with a split ratio of 20. An amount of 45 mL min⁻¹ H₂ and 450 mL min⁻¹ synthetic air were used as carrier gas. Separation of hydroxyalkanoic acid methyl esters occurred by using a temperature program (temperature hold for 4 min at 50 °C, temperature ramp with 40 °C per minute to a maximum of 240 °C, and temperature hold at 240 °C for 3 min) and esters were detected using flame ionization detector (FID). For calibration, poly-3-hydroxybutyrate from its natural origin (Sigma-Aldrich Chemie GmbH, Steinwehe, Germany) was processed to hydroxyalkanoic acid methyl esters as described above and used as an external quantification standard.

2.6. Visualization of Poly-3-Hydroxybutyrate In Vivo

2.6.1. Transmission Electron Microscopy

For scanning electron microscopy, the cells were immobilized on silicon platelets and fixed for 1 h using a final concentration of 2.5% glutaraldehyde in 0.1 mol phosphate buffer with 1% sucrose. Subsequently, the samples were postfixed with 2% osmium tetroxide in phosphate buffer for 20 min, dehydrated using a graded series of propanol, and then subjected to critical point drying using carbon dioxide in a CPD BalTec030 critical point dryer. Finally, the samples were mounted on specimen stubs and coated with approximately 2 nm of platinum using electron beam evaporation. The samples were imaged using a Hitachi S-5200 scanning electron microscope at an accelerating voltage of 10 kV, with the secondary electron signal being utilized. The samples were imaged using a Jeol-1400 transmission electron microscope at an accelerating voltage of 120 kV.

2.6.2. Fluorescence Microscopy

For fluorescence microscopy pictures, cells were harvested (5 min, 2500× g, 4 °C) and suspended in PBS. Afterward, cells were stained with 54 µM LipidGreen2 (in DMSO) for 10 min at room temperature and with light exclusion. After washing with PBS (5 min, 2500× g, 4 °C), cells were embedded in agarose (1% in 0.8% saline), and covered with a cover slide. Microscopic pictures were acquired using the ‘Zeiss AxioObserver.Z1 Microscope’ (Zeiss, Oberkochen, Germany) and a ‘Plan-Apochromat 63×/1.40 Oil Ph3 objective’ or ‘Plan-Apochromat 100×/1.40 Oil Ph3 objective’ (Zeiss, Oberkochen, Germany). All pictures were captured and processed using the same settings. For LipidGreen2, a specific excitation/emission wavelength of 440/505 nm was chosen, and the exposure time was set to 50 ms. The black/white contrast of brightfield pictures was set to 0/4096 and of fluorescence to 2.00/4096, and gamma values were set to 1.00.

3. Results

3.1. Heterotrophic Growth of Recombinant A. Woodii Strains

The plasmid p83_PHB_Scaceti was transformed into electrocompetent A. woodii cells. Production of poly-3-hydroxybutyrate with recombinant A. woodii [p83_PHB_Scaceti] was examined by the performance of a heterotrophic batch experiment using 60 mM fructose as a carbon source (Figure 1). The PHB-producer strain reached a maximal optical density (OD₆₀₀) of 4.2 ± 0.1 while consuming 55.6 ± 0 mM fructose. During heterotrophic growth, 138.5 ± 9.5 mM of acetate and 20.35% PHB/CDW were produced. Production of 598.3 mg PHB/L implies a yield of 0.06 g PHB/g fructose. The control strains A. woodii wild-type and A. woodii [p83] produced no PHB but solely acetate in concentrations of 147.1 ± 5.3 mM and 166.0 ± 6.0 mM, respectively. Fructose was consumed completely by both control strains resulting in a maximal optical density of A. woodii wild-type and A. woodii [p83] of 1.9 ± 0.1 and 2.6 ± 0, respectively. The pH dropped for all strains during growth similarly.

3.2. Establishment of a LipidGreen2 Assay

A standard method of PHB quantification includes lyophilization of biomass and acidic methanolysis, followed by GC analysis [12]. This method limits PHB quantification during bacterial fermentation due to this time-consuming procedure. Using the fluorescent dye LG2, which binds to neutral lipids only, PHB granules can be detected and thus PHB can be quantified during fermentation in a few minutes. Thereby, the best moment for cell harvest can be elucidated [9].

For the establishment of an LG2 assay appropriate for PHB quantification in recombinant A. woodii, optimal values for the parameters OD₆₀₀, LG2 concentration, and inclusion of washing steps were elucidated. First, OD₆₀₀ was set to 0.1, 0.5, and 1, and cells were incubated with 54 µM of LG2 for 10 min. Due to low background fluorescence, the clearest difference between A. woodii wild-type and A. woodii [p83_PHB_Scaceti] could be observed at OD₆₀₀ of 1. Next, the LG2 concentrations 10.8 µM, 21.6 µM, 54 µM, and 108 µM were tested while OD₆₀₀ was set to 1 and incubation time was kept at 10 min. All LG2 concentrations would have been suitable for fluorescence measurements, using comparability the concentration was kept at 54 µM. Independent of LG2 concentration, a moderate background fluorescence was noticed. To diminish the background fluorescence a washing step using PBS was integrated into the quantification protocol (data not shown).

During the heterotrophic growth of A. woodii [p83_PHB_Scaceti], fluorescence intensity was determined using the LG2 assay after 24, 48, 72, 96, and 168 h. It was observed that fluorescence intensity increased from 5897 ± 444 a. u. after 24 h to a maximum of 117,678 ± 5370 a. u. after 168 h of growth. However, the fluorescence intensity of the control strains A. woodii wild-type and A. woodii [p83] increased slightly from 2169 ± 328 a. u. to 8814 ± 655 a. u. and from 2410 ± 1116 a. u. to 3127 ± 240 a. u., respectively.

To obtain an overview of the correlation between fluorescence measurement using LG2 and PHB content in recombinant A. woodii cells, both parameters were visualized in Figure 2A,B. Thereby, it was obvious that fluorescence intensity and PHB content (per CDW) showed a linear correlation from 12% PHB/CDW and more. Below a PHB content of 12%/CDW, no correlation could be seen. Additionally, the Pearson correlation efficiency of 0.97 was calculated, which confirmed the linear correlation between PHB content and fluorescence intensity. As the aim of the establishment of an LG2 assay was the quantification of PHB content using fluorescence measurements only, a formula for the calculation of PHB content was developed. Therefore, variations of fluorescence intensity, e.g., log₂ of fluorescence intensity and fluorescence intensity squared, were tested. The highest correlation efficiency was reached using the log₂ of fluorescence intensity. PHB content can thus be quantified using Equation (1).

$$PHB \left[\%/CDW\right]=\frac{{\log }_{2}FL{U}_{440/505nm}-15.367}{0.0645}$$

(1)

3.3. Autotrophic Production of PHB

The recombinant PHB-producer A. woodii [p83_PHB_Scaceti] was further examined for PHB production using CO₂ + H₂ as carbon and electron source. During autotrophic growth, the recombinant strain reached a maximal OD₆₀₀ of 1.75 ± 0.07 while metabolizing 4.72 ± 0.10 bar of CO₂ + H₂ (Figure 3). A maximum of 161.16 ± 8.87 mM of acetate was produced. Measurement of fluorescence intensity using the LG2 assay resulted in a maximal fluorescence intensity of 33,349 ± 830 a. u. after 168 h decreasing to 21,032 ± 2226 a. u. after 1004 h of growth. However, PHB could not be detected after acidic methanolysis of the cells and subsequent GC analysis. Additionally, the calculation of PHB content using Equation (1) led to negative values, indicating that the equation could only be used for PHB values of 12%/CDW and higher. As the fluorescence intensity of the LipidGreen2-assay for A. woodii [p83_PHB_Scaceti] was significantly higher than that of the control strains, it was assumed that the produced amount of PHB was below the detection limit. Therefore, biomass samples of the triplicates were pooled for each time point (168, 504, and 1004 h). Furthermore, the volume of acidic methanol, chloroform, and water for acidic methanolysis was reduced by factor 10. Subsequently, GC analysis revealed the production of 1.92% PHB/CDW and 23.52 mg PHB/L after 168 h of autotrophic growth. PHB content decreased after 504 and 1004 h to 1.09 and 1.03% PHB/CDW, respectively. The control strains A. woodii wild-type and A. woodii [p83] reached a maximal OD₆₀₀ of 1.29 ± 0.04 and 1.68 ± 0.03, respectively. During autotrophic growth, the control strains used 4.88 ± 0.10 and 4.82 ± 0.29 bar of CO₂ + H₂ while producing 153.37 ± 5.23 mM and 152.44 ± 5.03 mM of acetate as sole product. There was no production of PHB. Fluorescence intensity measured using the LG2 assay increased for A. woodii wild-type and A. woodii [p83] from 1224 ± 64 a. u. to 5510 ± 1664 a. u. and from 1354 ± 58 a. u. to 2663 ± 80 a. u. The pH of all strains dropped similarly during growth from approximately 7.25–7.30 to 5.5–5.7.

3.4. Visualization of PHB Granula in Recombinant A. Woodii

Via transmission electron microscopy (TEM), it was visualized that the native PHB producers, e.g., Cupriavidus necator, accumulate PHB during imbalanced growth conditions in granules. The PHB granules of C. necator are evenly spread all over the cytoplasm and are alike in size [17]. PHB granules accumulating in recombinant C. coskatii or C. ljungdahlii cells differ both in size and distribution. Autotrophically grown recombinant cells contained PHB granules with a size between 0.5 µm and 2.2 µm, which resulted in a deformed cell shape because of the size of the PHB granules. It was also observed by Flüchter et al. that PHB production occurred only in some cells [6].

During heterotrophic fermentation using A. woodii [p83_PHB_Scaceti], heterogeneity in PHB production could be observed via transmission electron microscopy (Figure 4). After 72 h of growth, 16% of recombinant cells showed no PHB granules, 14% produced PHB granules spread all over the cytoplasm, and 70% of the cells showed accumulation of PHB granules at one or both poles of the cell. The size of PHB granules in one cell varies in size, but the deformation of the cell shape was not observed after 72 h of growth. The heterogeneity in PHB production was also noticed after 168 h of fermentation. An amount of 9% of recombinant cells produced no PHB granules, while 65% of cells showed PHB granules at one or both poles of the cell. In 26% of recombinant A. woodii [p83_PHB_Scaceti], the cytoplasm was filled with PHB granules of different sizes. Some cells showed deformation of the cell shape due to the amount and the size of the granules.

PHB production was also monitored via fluorescence microscopy during the heterotrophic growth of A. woodii [p83_PHB_Scaceti]. Cells were grown for 72 h with fructose as substrate and further stained with LG2. The fluorescence was detected at the poles of the cells, confirming the transmission electron microscopic pictures (Figure 5).

4. Discussion

Conventional plastics have undeniably contributed to societal advancements, but have also imposed severe environmental challenges. Thus, bioplastics offer a promising avenue for addressing these concerns by providing alternatives derived from renewable resources and offering improved end-of-life options. Among the promising group of PHAs, the biopolymer PHB is the most extensively studied substitute for conventional plastics [18]. In 1994, the thermoplastic PHB was the first isolated PHA with properties comparable to polypropylene (PP) regarding brittleness, crystallinity, and melting temperatures [19,20]. In contrast to PP or polyethylene (PE), PHB can be produced from renewable resources and is biodegradable. Additionally, it comprises a wide range of applications, e.g., in the car industry, packaging industry, or biomedical applications [21]. Furthermore, PHB is not only used in its pure form but also commercially available as a copolymer (e.g., with valerate) or as a blend with natural raw materials, which even improves the bioplastic’s properties for industrial usage [4]. PHB synthesis occurs in many different microbes during growth under imbalanced growth conditions and serves as an intracellular carbon and energy storage compound. Thus, PHB accumulates in storage granules [22]. C. necator, formerly known as Ralstonia eutropha or Alcaligenes eutrophus, serves as a model organism for PHB production via bacterial fermentation [17]. It forms PHB granules during imbalanced growth, implemented by carbon excess with nutrient limitations [23]. In the natural PHB producer C. necator, the genes for a β-ketothiolase, an acetyl-coenzyme A (CoA) reductase, and a polyhydroxyalkanoate synthase are responsible for PHB production (Figure 6) [8]. PHB production via bacterial fermentation using C. necator was investigated extensively regarding parameters such as medium composition, fermentation processes, and isolation of PHB [18,24]. Despite the narrow substrate spectrum of different strains of C. necator, fermentation with renewable resources or waste products, e. g. glycerol (C. necator DSM 545), different kinds of molasses (C. necator DSM 545, ATCC 25207) or C₁ gases (C. necator DSM 545, H16), is feasible [25,26]. Autotrophic production of PHB using C. necator seems very promising, due to a low-cost substrate, but implementation of the fermentation at industrial scale proved to be challenging. Growth conditions for obtaining biomass and production of PHB differ, which indicates that a two-step fermentation is necessary. Additionally, safety instructions for industrial-scale fermentation dictate lower O₂ concentrations in a gas mixture containing O₂ together with H₂, as necessary for efficient cultivation of C. necator. Thus, biomass production is limited with lower O₂ concentrations, which results in lower PHB yields as experienced under optimal growth conditions [27]. It must also be noted that PHB quality varies among different substrates and medium compositions, which manifests in diverse melting and glass transition temperatures, crystallinity, and brittleness of the biopolymer [18].

To exclude danger because of a mixture of CO₂, H₂, and O₂, autotrophic fermentation with recombinant acetogens using syngas (a mixture of CO₂, CO, and H₂) was investigated. The organism C. autoethanogenum, which is already used for industrial production of ethanol by LanzaTech, was transformed with codon-optimized genes of C. necator. Autotrophic growth of the recombinant C. autoethanogenum resulted in the production of 5.6% PHB/CDW [9]. The insertion of the native PHB-production genes of C. necator and Burkholderia thailandensis was also examined for C. ljungdahlii and C. coskatii, but no production could be observed. Additionally, recombinant PHB production using codon-optimized genes of C. necator was tried out, but plasmid construction was not successful. Thus, an alternative, synthetic PHB-production pathway was established. In C. acetireducens, a thiolase A catalyzes the formation of acetoacetyl-CoA from two moieties of acetyl-CoA. Acetoacetyl-CoA is reduced to (S)-3-hydroxybutyryl-CoA by a 3-hydroxybutyryl-CoA dehydrogenase. As polymerization of the S-isomer of 3-hydroxybutyryl-CoA is not possible, it is converted via a crotonase to crotonyl-CoA. Using the (R)-enoyl-CoA hydratase crotonyl-CoA can be hydrated, which results (R)-3-hydroxybutyryl-CoA. The R-isomer of 3-hydroxybutyryl-CoA can then be polymerized to PHB via the polyhydroxyalkanoate synthase (Figure 6) [6]. For recombinant PHB production in C. ljungdahlii and C. coskatii, the plasmid p83_PHB_Scaceti was constructed. It harbors the genes thlA, hbd, and crt (Locus tags: Ga0077986_111534, Ga0077986_111533, and Ga0077986_111532) originating from C. scatologenes, and encodes a thiolase A, a 3-hydroxybutyryl-CoA dehydrogenase and a crotonase, respectively. The genes phaJ and phaEC (Locus tags: CLOACE_21160, CLOACE_21150, and CLOACE_21140) originating from C. acetireducens encode a (R)-enoyl-CoA hydratase and a polyhydroxyalkanoate synthase, respectively. With the recombinant strains production of 1.2% PHB/CDW in C. coskatii and C. ljungdahlii was possible [6].

In this study, the synthetic PHB-production pathway of Flüchter et al. was inserted into the acetogen A. woodii, which serves as a model organism for acetogens [6]. Autotrophic growth was performed using CO₂ + H₂ as a substrate for the production of PHB, which resulted in a maximal yield of 1.9% PHB/CDW after 168 h of fermentation. This is the highest recombinant production of PHB with acetogens using the synthetic PHB-production pathway until now. Interestingly, PHB content decreased after 504 and 1004 h of fermentation, indicating that A. woodii possibly metabolizes PHB during autotrophic growth. Comparison of the amino acid sequence for the annotated PHB depolymerases of C. necator via BLAST analysis with the genome of A. woodii yielded no similar genes or protein sequences, respectively. Also, a comparison of the PHB depolymerases of C. necator with the amino acid sequences of the native, clostridial PHB-producers C. acetireducens, C. lundense, and C. tetanomorphum resulted in no significant similarity. Additionally, intracellular PHB degradation in clostridial PHB producers is not investigated yet and responsible genes are not annotated in the respective genomes. Thus, no published gene sequence for PHA depolymerases was found in the genome of A. woodii.

Production and degradation of PHB in A. woodii were also detected via fluorescence measurements using the newly established and modified LipidGreen2-assay [10]. Another challenge to overcome, besides optimization of fermentation conditions and isolation procedures, is the quantification of PHB. Nowadays, PHB is quantified using the quantification procedures of Brandl et al. or Karr et al., which are both very time-consuming and thus not efficient for industrial use [11,12]. Detection of PHB was also carried out using different fluorescent dyes, e.g., Nile red or BODIPY^493/503 [14]. Recently, the fluorescent dye LipidGreen2 was developed. LG2 is a small molecule probe, which binds selectively to neutral and not to polar lipids, thus reducing background fluorescence. It was initially developed for imaging of fatty liver in zebrafish, but lately, Kettner et al. performed fluorescence staining of PHB granules using LG2 [10,13]. In A. woodii, PHB granules were visualized using LG2 staining demonstrating that PHB granules accumulate during heterotrophic growth at one or both poles of the recombinant cells. During heterotrophic fermentation, some cells accumulate granules spread all over the cell, but most cells show the polar production of PHB granules. The heterogeneity regarding PHB production in recombinant cells was seen before in recombinant C. coskatii and C. ljungdahlii cells, but in a more distinct manner [6]. As it is known that different phasins are included in PHB formation and degradation in C. necator, a lack of phasins may cause heterogeneity in recombinant acetogens [28]. Phasins affect four major components of PHB metabolism, including PHB depolymerization, influence on PHA synthase activity, granule segregation, and chaperone activity. The activity of phasins is coordinated by a regulator phasin [29]. Interestingly, the annotated gene for the polyhydroxyalkanoate synthesis regulator phasin of the PHB producer C. lundense (locus tag: T514DRAFT_04392) can be found with varying similarities in the genomes of C. autoethanogenum, C. coskatii, C. ljungdahlii, and A. woodii (locus tags: CAETHG_2191, CLCOS_21310, CLJU_c00740, and Awo_c22930, respectively). In none of the native, clostridial producers’ genome a phasin is annotated. Additionally, there is no similarity between the phasins of C. necator and the genome of the clostridial PHB producers.

As this is the first time acetogenic production of PHB using CO₂ + H₂ was established and PHB yield is far from the maximum theoretical yield, further investigations for improvement must be considered. Due to low gas transition during low-scale batch-fermentation processes, transfer into a bioreactor and continuous sparging of the medium with gas can be advantageous. For the acetogen C. ljungdahlii production of ethanol could be increased by more than factor 10 by transfer into a bioreactor conducting continuous fermentation [30]. One aspect of continuous fermentation using A. woodii, which complicates the process, is high washout rates. To avoid the negative effects of washout of the cells, immobilization material, e.g., linen or polyethylene terephthalate (PET), can be integrated into the bioreactor promoting biofilm production [31]. Further, a bottleneck of the Wood–Ljungdahl pathway in acetogens is the supply of ATP and tetrahydrofolate (THF). Therefore, overexpression of the genes formyl-THF-synthetase, methenyl-THF-cyclohydrolase, methylene-THF-dehydrogenase, and methylene-THF-reductase in A. woodii leads to increased THF and acetate production [32]. It also enables A. woodii to cope with higher hydrogen partial pressure resulting in higher biomass and product concentrations [33]. For industrially scalable production of A. woodii [p83_PHB_Scaceti] using CO₂ + H₂, further investigation regarding gas transition into the medium and gas uptake, as well as overcoming the bottleneck of the Wood–Ljungdahl pathway, should be conducted.

For simplification of quantification processes, a linear correlation between fluorescence intensity and PHB content per CDW was detected. Amounts of more than 12% PHB/CDW can be calculated via the LG2 assay. As quantification of small amounts of PHB using GC analysis proved to be difficult, the quantification method using LG2 needs further investigation for quantification of low PHB levels. Although low PHB levels are not quantifiable yet, the LG2 assay is used to determine the time point for cell harvest and further processing. A reliable and easy quantification method could improve the economic competitiveness of PHB compared to conventional plastics, as fermentation optimizations can be evaluated during processing, and harvesting time points could be analyzed quickly [14].