Palm Oil is the highest produced edible oil globally, with over 66 million tonnes produced annually. It has been estimated that up to 50% of all products sold in the supermarket contain palm oil in some form. Palm oil has attractive properties such as a high melting point and texture due to a balanced ratio of unsaturated and saturated fatty acids. It contains approximately 40% oleic acid (monounsaturated fatty acid), 10% linoleic acid (polyunsaturated fatty acid), 45% palmitic acid and 5% stearic acid (saturated fatty acid), that results in an edible oil that is suitable for use in a variety of food, detergent and cosmetics products. In addition, palm oil is the least expensive oil produced due to its high productivity and extensive production. Due to the high demand for the product, vast amounts of rainforest have been cleared to make way for more plantations, reducing biodiversity and releasing huge levels of carbon dioxide into the atmosphere. There is a clear need for an alternative lipid that can match palm oils properties but can be produced sustainably. Recent work suggests that some yeasts are capable of producing a similar oil to palm oil and can be grown on waste resources. In this thesis a novel bioprospecting protocol was developed to isolate yeasts that can survive the harsh conditions necessary for industrial biotechnology. In this way a vineyard and the local area was sampled for yeasts which were then cultured under extremes of pH, multiple sugars and inhibitors caused from the breakdown of lignocellulose. The wild yeast were cultured in four stages: minimal medium with Lysine; minimal medium with inhibitors; minimal medium with xylose as sole carbon-source; and lastly minimal medium with only arabinose and cellobiose as carbon-sources. Only strains that survived each stage were taken forward to the next, to isolate species that were truly suited to these conditions. Out of the estimated 1000s of strains screened this resulted in 12 strains of yeast, mostly in the Metschnikowia pulcherrima, group being able to cope with the conditions. The 12 strains were further analyzed by culturing them in an array of 4 different model lignocellulosic feedstocks namely wheat straw, corn Stover, sugarcane bagasse, and palm kernel cake hydrolysates. Other conditions incorporated in these analysis were a range of pH from pH 1.5 to pH 7.0; four levels of a mixture of 5 inhibitors; and two different temperatures. All of the 12 strains showed similar behaviour where inhibitor tolerance was only marked at higher pH, and at low pH the strains could not grow at all. Though all strains were able to grow on the hydrolysate models, even those with little glucose and/or xylose content. The lipid profile of the strains was also assessed and proved to be similar to most terrestrial crops, with suitable lipid profiles for a rapeseed oil, and in some cases palm oil substitute. Lastly, to further evaluate the accurate identification of the strains as there are some ambiguity in the Metschnikowia pulcherrima group, we applied an approach only widely used for Pathogenic Bacteria/Yeast identification, Multilocus Sequence Typing (MLST). Using 25 strains (7 of this collection), 6 type species and some isolates from the original culture collection in Bath. Sequences of 6 genes was analysed using the Bayesian statistical method. The result showed grouping of M. pulcherrima into 3-4 groups 9 different for each gene. M. Corniflorae being the outgroup. In all 3 genes successfully sequenced: M. Fruticola; R6; Mp DAH 3; and ICS48 were consistently shown to be clonal. The work presented here demonstrates a new method for bioprospecting strains capable of isolating strains for industrial biotechnology, and for characterisation of the yeast in the Metschnikowia genus. Some of the yeasts identified were oleaginous, and could potentially be used as a novel source of palm oil substitute.

A exploração de resíduos agroindustriais como matérias-primas aplicadas na conversão de lipídeos por leveduras, possibilitam um destino mais sustentável a estes resíduos, visto que matéria-prima lipídica possui grande interesse comercial, tanto para suplementação alimentar como na síntese de biocombustíveis e outros produtos da indústria oleoquímica. Foram avaliadas nove linhagens de leveduras, identificadas anteriormente como boas ou ótimas produtoras de lipídeos em meio mel, para seleção da estirpe mais adequada para a produção de lipídeos em meio de soro de queijo. O crescimento celular foi quantificado pela determinação gravimétrica da biomassa seca a 60ºC por 24 h e a extração dos lipídeos totais foi determinada utilizando o método de Bligh e Dyer. A maior produção de lipídeos totais foi de 1,27 g L-1 obtida pela linhagem de Cryptococcus laurentii, apresentando diferença significativa em relação às demais linhagens avaliadas, pelo teste de Tukey a 5% de significância. Posteriormente foi realizada a comparação de dois meios de cultivos e dois métodos de extração de lipídeos de cultura de C. laurentii. Os experimentos foram realizados com planejamento fatorial completo 22, com dois fatores e dois níveis. Os lipídeos foram extraídos em dois diferentes métodos: Bligh e Dyer e Folch et al.; as leveduras cultivadas em dois meios de cultivo: soro de queijo e YEPG líquido. Obteve-se conteúdo lipídico final superior em soro de queijo nas condições avaliadas, já o método de extração de Bligh e Dyer não apresentou diferenças estatísticas em comparação ao de Folch et al., contudo apresentou maiores médias, sendo adotado como método de extração dos experimentos seguintes. Realizou-se um estudo para otimização da produção de biomassa e lipídeos totais de C. laurentii, em meio de cultivo de soro de queijo suplementado com melaço de cana de açúcar. A primeira etapa consistiu no estudo do efeito do pH, tempo de fermentação e concentração de melaço. A segunda etapa consistiu no efeito da suplementação por extrato de levedura e sais inorgânicos (KH2PO4, Na2HPO4, MgSO47H2O, CaCl22H2O, FeCl36H2O e ZnSO4H2O). A condição ótima de cultivo para linhagem de C. laurentii em meio de soro de queijo foi definida com os seguintes fatores: 360 h de fermentação, pH do meio de cultivo de 6,5 e suplementação (g L-1): 50 melaço, 0,5 extrato de levedura, 4 K2PO4, 1 Na2PO4, 0,75 MgSO47H2O e 0,002 ZnSO4H2O. Os lipídeos totais produzidos na condição de cultivo alcançaram 2,96 g L-1, que representa um aumento de 133,1% na produção de lipídeos totais por C. laurentii em relação ao meio com soro de queijo sem suplementação. Ocorrendo predomínio de lipídeos neutros 85,76%, com alto teor de ácidos graxos saturados e monosaturados, de cadeia longa com 16 e 18 átomos de carbono, o que possibilita sua aplicabilidade como fonte de triglicerídeos para biodiesel. / The exploitation of agro-industrial wastes as raw materials applied in the conversion of lipids by yeasts, allows a more sustainable destination of such waste, since lipid feedstock has great commercial interest, both for food supplementation as in the synthesis of biofuels and other products for oleochemical industry. We have evaluated nine strains of yeast, previously identified as good or excellent producers of lipids in honey medium, for selecting the most suitable strain for the production of lipids in cheese whey medium. Cell growth was quantified by gravimetric determination of dry biomass at 60°C for 24 h and the extraction of total lipids was determined using the method of Bligh and Dyer. The highest yield of total lipids was 1.27 g L-1 produced by the strain of Cryptococcus laurentii, significant difference compared to other strains assessed by Tukey test at 5% significance level. Posteriorly was carried out a comparison of two culture media and two lipid extraction methods for culture of C. laurentii. The experiments were performed with 22 full factorial design used two factors and two levels. Lipids were extracted by two different methods: Bligh and Dyer and Folch et al.; Yeasts grown in two culture media: cheese whey and liquid YEPG. Was achieved lipid content upper in cheese whey under the conditions evaluated, since the extraction method of Bligh and Dyer showed no statistical differences compared to that of Folch et al., but had higher average, used as a method of extraction in following experiments. Was conducted a study to optimize the production of biomass and lipid content of C. laurentii, in culture medium of whey supplemented with sugar cane molasses. The first step was to study the effect of pH, fermentation time and molasses concentration. The second step was the effect of supplementation by yeast extract and inorganic salts (KH2PO4, Na2HPO4, MgSO47H2O, CaCl22H2O, FeCl36H2O, and ZnSO4H2O). The optimum condition for growing strain of C. laurentii in the cheese whey medium was set to the following factors: 360 h fermentation, 6,5 pH and supplementation of (g L-1): 50 molasses, 0.5, yeast extract, 4 K2PO4,1 Na2PO4, 0.75 MgSO47H2O and 0.002 ZnSO4H2O. The total lipid produced in the culture condition reached 2.96 g L-1, which represents an increase of 133.1% total lipid in the production of C. laurentii in relation to the cheese whey medium without supplementation. Predominantly occurred neutral lipids 85.76%, with high content of saturated and monosaturated fatty acids, long chain with 16 and 18 carbon atoms, which makes its applicability as a source of triglycerides for biodiesel

A exploração de resíduos agroindustriais como matérias-primas aplicadas na conversão de lipídeos por leveduras, possibilitam um destino mais sustentável a estes resíduos, visto que matéria-prima lipídica possui grande interesse comercial, tanto para suplementação alimentar como na síntese de biocombustíveis e outros produtos da indústria oleoquímica. Foram avaliadas nove linhagens de leveduras, identificadas anteriormente como boas ou ótimas produtoras de lipídeos em meio mel, para seleção da estirpe mais adequada para a produção de lipídeos em meio de soro de queijo. O crescimento celular foi quantificado pela determinação gravimétrica da biomassa seca a 60ºC por 24 h e a extração dos lipídeos totais foi determinada utilizando o método de Bligh e Dyer. A maior produção de lipídeos totais foi de 1,27 g L-1 obtida pela linhagem de Cryptococcus laurentii, apresentando diferença significativa em relação às demais linhagens avaliadas, pelo teste de Tukey a 5% de significância. Posteriormente foi realizada a comparação de dois meios de cultivos e dois métodos de extração de lipídeos de cultura de C. laurentii. Os experimentos foram realizados com planejamento fatorial completo 22, com dois fatores e dois níveis. Os lipídeos foram extraídos em dois diferentes métodos: Bligh e Dyer e Folch et al.; as leveduras cultivadas em dois meios de cultivo: soro de queijo e YEPG líquido. Obteve-se conteúdo lipídico final superior em soro de queijo nas condições avaliadas, já o método de extração de Bligh e Dyer não apresentou diferenças estatísticas em comparação ao de Folch et al., contudo apresentou maiores médias, sendo adotado como método de extração dos experimentos seguintes. Realizou-se um estudo para otimização da produção de biomassa e lipídeos totais de C. laurentii, em meio de cultivo de soro de queijo suplementado com melaço de cana de açúcar. A primeira etapa consistiu no estudo do efeito do pH, tempo de fermentação e concentração de melaço. A segunda etapa consistiu no efeito da suplementação por extrato de levedura e sais inorgânicos (KH2PO4, Na2HPO4, MgSO47H2O, CaCl22H2O, FeCl36H2O e ZnSO4H2O). A condição ótima de cultivo para linhagem de C. laurentii em meio de soro de queijo foi definida com os seguintes fatores: 360 h de fermentação, pH do meio de cultivo de 6,5 e suplementação (g L-1): 50 melaço, 0,5 extrato de levedura, 4 K2PO4, 1 Na2PO4, 0,75 MgSO47H2O e 0,002 ZnSO4H2O. Os lipídeos totais produzidos na condição de cultivo alcançaram 2,96 g L-1, que representa um aumento de 133,1% na produção de lipídeos totais por C. laurentii em relação ao meio com soro de queijo sem suplementação. Ocorrendo predomínio de lipídeos neutros 85,76%, com alto teor de ácidos graxos saturados e monosaturados, de cadeia longa com 16 e 18 átomos de carbono, o que possibilita sua aplicabilidade como fonte de triglicerídeos para biodiesel. / The exploitation of agro-industrial wastes as raw materials applied in the conversion of lipids by yeasts, allows a more sustainable destination of such waste, since lipid feedstock has great commercial interest, both for food supplementation as in the synthesis of biofuels and other products for oleochemical industry. We have evaluated nine strains of yeast, previously identified as good or excellent producers of lipids in honey medium, for selecting the most suitable strain for the production of lipids in cheese whey medium. Cell growth was quantified by gravimetric determination of dry biomass at 60°C for 24 h and the extraction of total lipids was determined using the method of Bligh and Dyer. The highest yield of total lipids was 1.27 g L-1 produced by the strain of Cryptococcus laurentii, significant difference compared to other strains assessed by Tukey test at 5% significance level. Posteriorly was carried out a comparison of two culture media and two lipid extraction methods for culture of C. laurentii. The experiments were performed with 22 full factorial design used two factors and two levels. Lipids were extracted by two different methods: Bligh and Dyer and Folch et al.; Yeasts grown in two culture media: cheese whey and liquid YEPG. Was achieved lipid content upper in cheese whey under the conditions evaluated, since the extraction method of Bligh and Dyer showed no statistical differences compared to that of Folch et al., but had higher average, used as a method of extraction in following experiments. Was conducted a study to optimize the production of biomass and lipid content of C. laurentii, in culture medium of whey supplemented with sugar cane molasses. The first step was to study the effect of pH, fermentation time and molasses concentration. The second step was the effect of supplementation by yeast extract and inorganic salts (KH2PO4, Na2HPO4, MgSO47H2O, CaCl22H2O, FeCl36H2O, and ZnSO4H2O). The optimum condition for growing strain of C. laurentii in the cheese whey medium was set to the following factors: 360 h fermentation, 6,5 pH and supplementation of (g L-1): 50 molasses, 0.5, yeast extract, 4 K2PO4,1 Na2PO4, 0.75 MgSO47H2O and 0.002 ZnSO4H2O. The total lipid produced in the culture condition reached 2.96 g L-1, which represents an increase of 133.1% total lipid in the production of C. laurentii in relation to the cheese whey medium without supplementation. Predominantly occurred neutral lipids 85.76%, with high content of saturated and monosaturated fatty acids, long chain with 16 and 18 carbon atoms, which makes its applicability as a source of triglycerides for biodiesel