Search

Polyhydroxyalkanoates (PHAs) are biopolymers that have the potential to replace oil-derived plastics, as they are entirely biodegradable and are microbially synthesised from renewable bio-based feedstocks. However, the broader uptake of PHAs is currently being prevented by their high production costs, inconsistent material properties, and processing challenges. Chief among these issues is the cost of production, of which the carbon substrate represents a significant portion (28 – 50 %). Hence, there is substantial interest in using cheap and readily available feedstocks, such as plant oils, biodiesel by-products, and other lipid substrates. Oil extracted from spent coffee grounds – a cheap and readily available feedstock generally discarded as waste – has been demonstrated to be a highly suitable feedstock able to achieve high biopolymer yields. Throughout this thesis, Cupriavidus necator DSM 545 – a bacterium widely renowned for its ability to accumulate high levels of PHAs (= 90 %, w/w) from a wide array of carbon substrates – was cultivated for the first time on SCG oil and other plant oilsto produce poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (P(3HB-co-3HV)) copolymers. The overall aim of the research is to improve PHA production by evaluating and optimising the different factors affecting the consumption of these lipid substrates in submerged fermentations. This thesis is presented in Journal (Alternative) Format and is split over three significant areas of study, presented as separate research articles. Initially, SCG oil was benchmarked against sunflower oil for cell growth and biopolymer production in shake flask and bioreactor (3 L) fermentations. On SCG oil, C. necator DSM 545 produced 8.9 g L-1 of total biomass, of which 89.6 % (w/w) was P(3HB-co-3HV) (6.8 mol% 3HV). In contrast, cells cultivated on sunflower oil reached 9.4 g L-1 of total biomass, with a P(3HB-co-3HV) (0.2 mol% 3HV) content fraction of 88.1 % (w/w). The production of 3HV monomers in the absence of traditional precursors is unusual and was ascribed to the activation of the succinate-propionate metabolic pathway. In doing so, the organism can divert succinyl-CoA away from a dysfunctional tricarboxylic acid (TCA) cycle into propionyl-CoA, which is a 3HV precursor. Including 3HV monomers into the biopolymer structure improves its material properties and thermal processability compared to pure P(3HB). Overall, the ability to synthesise 3HV monomers from a structurally unrelated substrate opens up the potential of eliminating the need for expensive precursor substrates typically required for copolymer synthesis, which will help to significantly reduce production costs. Second, the range of evaluated substrates was expanded to include olive, rapeseed, and sesame seed oils, and the influence of fatty and organic elemental composition on the PHA yields was assessed. During PHA production, the organism was observed to preferentially consume saturated fatty acids (palmitic acid in particular). By analysing the substrate preferences of an organism, it is possible to select the most appropriate substrates for fermentation, including those which are often neglected in industrial use, including SCGs. Among the evaluated plant oils, SCG oil produced the highest concentrations of total biomass (11.3 g L-1 ) and P(3HB-co-3HV) (10.0 ± 0.6 g L-1 ; 3.7 mol% 3HV), which was due to the high acid value and palmitic acid content of the oil. The corresponding polymer fractions (as high as 88.9 % (w/w)) are among the highest ever reported values in the literature. As plant oils are immiscible in aqueous fermentation media, they tend to float at the surface, remaining largely out of the organism's reach. Emulsifying the oils with a surfactant (dimethyl sulfoxide) reduced the duration of the initial lag phase, which can help increase PHA volumetric productivity. Finally, the growth and PHA production of C. necator DSM 545 from SCG oil were optimised by varying the evaluating the relationship between the specific growth rate (µ) and the initial substrate (carbon) concentrations (C0). Saponifying the oil was also studied for the first time as an alternative to emulsification to increase its bioavailability. In preliminary shake flask experiments, values of µ peaked at 0.139 h-1 and 0.145 h -1 on the crude and saponified SCG oil, respectively, when C0 = 8.6 g L-1 . Cell growth was completely inhibited in the saponified SCG oil flasks when C0 reached 19.2 g L-1 . The relationship between µ and C0 was further analysed by curve-fitting several Monod-based kinetic growth models to the experimental data; improving the understanding of this relationship is imperative to developing suitable feeding regimes. Biomass production was subsequently increased by mixing the crude and saponified substrates together in a carbon ratio of 75:25% (w/w), respectively – such that the SCG oil effectively became its own surfactant. In bioreactors, cells grew 52 % faster on the mixed substrates (µ = 0.35 h -1 ) than on crude SCG oil (µ = 0.23 h -1 ). After 72 h, cells grew to 8.5 g/L of total biomass and collected 84.4 % (w/w) of P(3HB) on the mixed substrates. In contrast, 7.8 g L-1 of total biomass and 77.8 % (w/w) P(3HB) were obtained on just SCG oil. Hence, saponification can be seen to increase both the growth rates and the final PHA concentrations. This PhD greatly expands on the existing body of literature by focussing on improving and better understanding how the carbon substrate is utilised. Using either saponification or emulsification has been shown to increase biopolymer productivity by improving substrate bioavailability in submerged fermentations. Further, by knowing the fatty acid preferences of the organism and the ideal and limiting concentrations of the carbon substrate, it is now possible to optimise feeding regimes for maximum PHA production. Overall, this work will help produce PHAs with suitable material properties at lower costs, thereby improving their commercial viability.

The development of human society, which is highly dependent on fossil fuels, is now facing a range of global issues, such as rising energy prices, energy security and climate changes. To successfully tackle the resultant issues, the energy transition from fossil fuels to renewable energy sources, such as solar energy, tide energy, hydroelectric power, geothermal heat and biofuels, is under way. Biodiesel, as an important type of biofuels, has been increasingly produced from vegetable oil or used cooking oil, especially in Europe. Nevertheless, considering the high production cost of biodiesel, there is still much to be done to improve the economics of biodiesel industry. Utilisation of crude glycerol, the main by-product of the biodiesel industry, to produce value-added products appears to be a promising solution. Poly(3-hydroxybutyric acid) (PHB), a biodegradable plastic, can be converted from glycerol by Cupriavidus necator DSM 545 under unbalanced growth conditions, such as nitrogen limitation. One way to enhance the batch production of PHB is to genetically engineer the strain of C. necator, which requires insights of the dynamic impact of extracellular environment on cell phenotypes. Hence in this thesis, we aim to perform metabolic modelling based on experimental measurements to gain a better understanding of the behaviour of the metabolic network of Cupriavidus necator DSM 545 and identify potential bottlenecks of the process. Initially, C. necator DSM 545 is a strain that hardly grows on glycerol, so in a preliminary study, we investigate the process by which the strain was adapted to consume glycerol through serial subcultivation. It is found that the adaptation can be achieved within 15 cell generations over three passages in basal mineral medium, and the acquired phenotype is sufficiently stable upon further passage. The study of metabolism started with the reconstruction of the cell's metabolic network, followed by a thermodynamic analysis to check the feasibility and reversibility of all the biochemical reactions included. Then the static flux balance analysis was extended and applied to analyse the shift of metabolic states during the microbial fermentation in different batch conditions. The resulting patterns of flux distribution reveal the TCA cycle to be the major competitor for PHB synthesis at the ACCoA node. Cells have the potential to enter an efficient PHB-production phase that features minimal TCA/PHB flux split ratio, and the length of the phase can be manipulated by aeration. Although low aeration rate favours optimal flux split ratio, such condition that limits respiration also limits nutrient uptake, leading to low PHB productivity overall. To identify the actual limiting factors of PHB synthesis in the system, we further performed metabolic control analysis based on the calculated flux distributions. The analysis demonstrated how the distribution of the metabolic control can vary widely, depending on the aeration conditions used and the flux split ratios. Glycerolipid pathway, glycolysis, PHB metabolism, as well as the electron transport chain are revealed to be potential engineering targets as they contribute to the great majority of the positive control of PHB flux.

To mitigate the problems of depleting and soaring price of fossil fuels, the production and use of renewable energy have been vigorously promoted. In Europe, the role of biologically-derived fuels and in particular biodiesel is gradually increasing in prominent. Rapeseed biodiesel is the most widely produced in Europe. As a consequence, enormous amount of by-products from production processes are being generated. Current strategies for managing these by-products (mainly rapeseed meal and crude glycerol) seem not to be economically sustainable. More efficient utilisation could add more value to the production chain which in turn would raise the competitiveness of biodiesel compared to petro-diesel. The aim of the project reported in this thesis was to study the feasibility of producing a value added product, polyhydroxybutyrate (PHB), from by-products generated from rapeseed biodiesel production processes as well as to investigate the effects of methanol, a major impurity in crude glycerol, on growth of Cupriavidus necator, a PHB-producing micro-organism.The preliminary study of C. necator growth in crude glycerol based media revealed that optimum concentration of crude glycerol was in a range 15-25 g/L. It was also found that slight changes in the carbon to nitrogen ratio of the feedstock did not significantly affect the growth while methanol at concentrations beyond 10 g/L did. A model based on a saturation equation was developed and used to successfully predict the inhibition of growth by methanol. From the developed model, mechanisms of the inhibition were proposed. The model could also be used to predict satisfactorily growth or productivity rates in other systems containing short-chain alcohols. The growth in solutions derived from rapeseed meal (designated as hydrolysate) via solid-state fermentation by Aspergillus oryzae followed by hydrolysis of the fermented solids was also studied. The biomass production was found to increase as a function of initial free amino nitrogen (FAN) concentration presented in the hydrolysate. However, at higher initial FAN concentrations, a lower conversion of nitrogen to biomass was observed. PHB production was studied using a feedstock which was a mixture of the hydrolysate and crude glycerol. Total biomass concentration reached 28.8 g/L at 120 h with 86% PHB content. PHB productivity and PHB yield on glycerol were 0.21 g/L•h and 0.32 g/g respectively. These results were comparable with those obtained when pure glycerol and synthetic crude glycerol were used, suggesting that, technically, the use of the generic rapeseed- and crude glycerol-based feedstock to produce PHB is feasible.Overall, the feasibility of producing PHB from rapeseed biodiesel by-products has been demonstrated. The satisfactory result leads to the more important outlook that the generic feedstock derived from rapeseed biodiesel by-products has the potential to be used to produce a wide range of products depending on the micro-organism used. Further development of this process to improve nutrient production efficiency as well as product yields and subsequent integration of the process into the biodiesel production process could well be an important contribution in the development of a sustainable biodiesel industry.

Synthesis gas (syngas) is an industrially important feed stock for electricity, hydrogen and liquid fuel production. Biomass-derived syngas has high CO2 levels and alternative methods of CO2 removal are needed to implicate biomass gasification systems. This research identified and tested a possible method of biological CO2 fixation from syngas. The chemolithoautotrophic organism Cupriavidus necator produces PHB and offers a valuable product as a result of CO2 fixation. The energy efficiency of biological CO2 fixation using chemolithoautotrophic growth supported by hydrogen, oxygen and carbon dioxide was theoretically analyzed and then tested in C. necator using a novel experiment. The experimental design included cultivating C. necator in a mineral solution inside of plastic gas sampling bags under constant temperature and pressure. The gas composition of the bag was monitored over time using a GC/TCD system. To determine the consumption of the individual gasses H2, O2 and CO2 the total moles of the gas mixture in the bag was monitored using the inert gas methane (CH4). The efficiency of chemolithoautotrophic growth was determined from gas consumption. It was found that there was a large discrepancy between the maximum theoretical energy efficiency and the actual efficiency. It was found that the gas consumption requirements of C. necator are 1:1.8:5.8 (mol CO2 : mol O2 : mol H2). This hydrogen requirement is almost three times more than the theoretically possible, and results in less than 30% of hydrogen consumed for growth being stored as reduced CO2, or bacterial biomass.