Estudo da produção de etanol a partir de xilose e xilooligômeros usando xilanases, xilose isomerase e levedura co-imobilizadas

Abstract

The production of ethanol from xylose with the yeast Saccharomyces cerevisiae was initiated in the group with the SIF process (simultaneous isomerization and fermentation). This process allows total conversion of xylose, but has been shown to be vulnerable to bacterial contamination. The process of simultaneous hydrolysis, isomerization and fermentation (SHIF) of xylooligomers was then proposed as an alternative for the fermentation of pentoses by commercial yeast. In this process, a xylanase cocktail hydrolyses the substrate to xylose, which, by the action of xylose isomerase, is transformed into xylulose, which is then fermented to ethanol by S.cerevisiae. All reactions take place within the biocatalyst where enzymes and yeast are co-immobilized. It was possible to produce ethanol by this process, but the first tests showed that the substrate conversion was incomplete, requiring investigation of the possible causes of the problem. The inefficiency of the hydrolysis stage, catalyzed by endo- / exoxylanases and β-xylosidase, a non-existent step in the SIF process, was the first hypothesis verified. The need to produce a standard substrate for the study of this reaction stage was the first challenge faced, since birchwood xylan, a commercial substrate previously used, was discontinued. A substrate was then produced in the laboratory from bleached eucalyptus pulp. Xylan was extracted with 4% NaOH solution at 25 °C, precipitated with glacial acetic acid and lyophilized. An autoclave heat pretreatment (15 min, 121 °C, 1 atm) allowed to increase the solubility of the carbohydrate chains. The xylan thus obtained was hydrolyzed by xylanases at the same rate as the commercial birchwood xylan, and made it possible to begin the investigation of the hydrolysis reaction. The observed xylobiose accumulation indicated that the inefficiency of the hydrolysis was due to the low concentration of β-xylosidase in the commercial enzymatic complexes. After cloning the Bacillus subtilis β-xylosidase gene in Escherichia coli, the recombinant enzyme was produced, purified and characterized. The results showed reduced enzyme activity at the process pH and low operational stability. The immobilization and stabilization of β-xylosidase in agarose-glyoxyl support proved to be efficient, as it considerably increased the thermal stability of the enzyme at 35 °C (164x) and 50 °C (3605x), with no significant loss of activity of the derivative after 10 consecutive hydrolysis cycles. The commercial xylanolytic complex Multifect, mainly responsible for the endoxylanase action, was immobilized on chitosan-glutaraldehyde. This technique does not allow a significant increase in the stability of complex enzymes with immobilization, but since soluble xylanases already have good stability at 35 °C, no new protocols have been tested. The obtained derivatives presented recovered activities maximum of 50%. Activity assays at different temperatures did not indicate the presence of diffusive effects, possibly being the enzymatic distortion due to the formation of multiple bonds with the support a probable reason for the 50% loss of immobilization activity. The obtained derivative was capable to hydrolyze the hydrothermal liquor in smaller XOS (X<6), with XOS yields of 56.3% and xylose of 12.9%. The high concentration of xylobiose (28.8 g.L-1) accumulated indicated a need to complement the enzymatic cocktail with more β-xylosidase. Hydrolyses supplemented with β-xylosidase derivative increased yields of xylose to 34.2%, yielding 46.5 g.L-1 xylose and reducing xylobiose to 7.4 g.L-1. The failure to obtain complete conversion of xyloligomers into xylose may be due to inhibitory effects of the product or limitation of the action of endoxylanases by their specificity to the substrate. The results obtained, however, were considered satisfactory for initial studies of the SHIF process. Another possible explanation for the SHIF halt would be the inhibitory effects of calcium ions and / or the hydrolysis reaction byproducts (e.g. XOS) on the action of xylose isomerase. Ca2+ isomerization tests showed that the reduction of the isomerization reaction rate truly occurs, with competition between the Ca2+ and Mg2+ ions for the metal site inside the enzyme. The isomerization velocity was affected by X2 only at high substrate concentrations (CS> 50 g.L-1), characterizing the competitive inhibition (R² = 0.99). The isomerization rate was significantly reduced under SHIF conditions (pH 5.6 35 ° C) and was even more affected by the combined effect of the presence of Ca2+ and X2 ions, although the reaction did not completely stop. These effects should therefore not be responsible for the incomplete conversion of the substrate. New assays of the SHIF process confirmed the importance of the β-xylosidase derivative in the biocatalyst composition, with an increase in ethanol production. The addition of more yeast to the reactor increased the consumption of the pentoses and led to an increase in ethanol production to 0.221 g.g-1 and 0.153 g.L-1.h-1), indicating that the SHIF process stagnation could also be related to the reduced speed of xylulose fermentation. It was decided at this point to compare the performance of genetically modified S. cerevisiae with that of commercial yeast in the fermentation of xylooligomers obtained by hydrothermal treatment of sugarcane bagasse. The SHIF process again showed stagnation, reaching only 5 g.L-1 of ethanol. Using the recombinant yeast, the simultaneous hydrolysis and fermentation-SHF process is performed, since this microorganism does the in vivo isomerization. In this process, ethanol concentration was increased to 15 g.L-1, generated from xylose already present in the hydrothermal liquor and the partial hydrolysis of xylooligomers. However, with the recombinant yeast stagnation was also observed, with accumulation of xylose in solution. These results thus indicate that the hydrolysis process generates some condition affecting the fermentation. The pH control allowed a small increase of the conversion in the SHIF and SHF processes, but did not prevent the stagnation. After the tests, 95% of the wild yeast and 50% of the genetically modified yeast remained viable. It has been observed that the reduction in fermentation rate may be related to an observed increase in acetic acid concentration generated in the hydrolysis step and which would be affecting both commercial and recombinant yeast. Thus, although it has not achieved full elucidation of the phenomena present in this complex process, this work allowed a great increase in the knowledge of the SHIF process, providing important indications for the necessary continuity of its study, in view of the total conversion of xylooligomers.