Les milieux biologiques bruts, provenant des opérations de transformation de biomasse sont souvent caractérisés à la fois par leur caractère polluant et par un potentiel de valorisation important. Le développement de procédés adaptés au traitement de tels milieux complexes présente ainsi beaucoup d’intérêt. Les jus verts générés par la déshydratation de la luzerne (Medicago Sativa) sont caractérisés par une forte teneur en protéines. Outre leur valeur nutritionnelle importante, ces protéines ont des applications potentielles dans plusieurs domaines, notamment environnemental de part leur teneur élevée en Ribulose 1,5 Bisphosphate Carboxylase Oxydase (Rubisco), enzyme responsable de la fixation du CO2 chez les plantes. Dans ce travail la séparation sélective de la Rubisco à partir du jus de luzerne industriel centrifugé par chromatographie d’échange d’ions a été étudiée. Dans un premier temps une méthode d’analyse qualitative et quantitative a été mise au point pour la détection et la quantification de la Rubisco en solution et ainsi le suivi du procédé de séparation. Dans un deuxième temps, le procédé de séparation a été étudié en colonne, en lit fixe et en lit expansé, en utilisant le support échangeur d’anions Q Hyper Z et l’effet de la dilution du milieu sur la capacité dynamique du procédé a été analysé dans les deux cas. Les résultats obtenus ont montré que les deux modes de contact permettent d’avoir des capacités dynamiques de rétention du même ordre de grandeur que celles de la littérature. Après élution, la Rubisco a été concentrée jusqu’à 21 fois et les fractions produites étaient caractérisées par un grand degré de pureté. Par ailleurs, des études d’équilibre et cinétique d’échange ont été initiées dans ce travail et ont démontré que malgré la taille importante de la protéine d’intérêt (560 kDa) les limitations stériques à son transfert ne sont pas plus importantes que dans le cas de protéines plus simples et plus petites et que le support Q Hyper Z présente effectivement une grande affinité pour la protéine. Enfin une première approche théorique a été conduite pour la compréhension des interactions entre la protéine et l’échangeur dans ce milieu complexe. Elle a permis de confirmer l’importance de la prise en compte de la présence d’autres biomolécules dans le milieu sur la rétention de la Rubisco, peut être même plus que celle des sels / Biological raw material derived from bio-refinery processes, is often considered a source of pollution but it seems also to be a promising alternative to potential material recovery. The development of suitable processes for handling such complex biological material has so many concerns. Green juice produced from mechanical dehydration of Alfalfa (Medicago sativa) is an excellent source of protein with high nutritional quality. Ribulose-1,5-bisphosphate carboxylase oxygenase (Rubisco) is the most abundant protein in the green juice, with potential applications in many fields, such as human nutrition, pharmaceuticals, environmental… The aim of this study is to isolate and recover Rubisco produced from an industrial alfalfa green juice, by ion exchange chromatography process. First of all, a qualitative and quantitative analytical method was developed to provide reliable information about Rubisco content monitoring in the separation process. In a second step, the separation process was performed in fixed and expanded bed, using the anion exchanger Q Hyper Z. In both cases, the effect of the dilution of the green juice on the dynamic capacity of the columns was studied. The results showed that the dynamic capacity retention was similar in both columns to those reported in literature. After elution step, Rubisco was concentrated 21 times and produced with high level of purity. Furthermore, kinetic of ion exchange study was initiated. Despite the large size of the protein (560 kDa), steric limitations to mass transfer were not very significant when compared to those of conventional small proteins. The support Q Hyper Z showed an excellent affinity for the protein recovery. Finally, a first theoretical investigation has been conducted for understanding the retention mechanism between the protein and the separation column. This study shows the importance of taking into account the presence of other bio-molecules in order to perform the retention of Rubisco, perhaps even more than that of salts

Made available in DSpace on 2014-12-17T15:01:34Z (GMT). No. of bitstreams: 1 CarlosEAP_DISSERT.pdf: 1904684 bytes, checksum: 4fd2147b17a381ad69d921436b5c83de (MD5) Previous issue date: 2013-12-18 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / Expanded Bed Adsorption (EBA) is an integrative process that combines concepts of chromatography and fluidization of solids. The many parameters involved and their synergistic effects complicate the optimization of the process. Fortunately, some mathematical tools have been developed in order to guide the investigation of the EBA system. In this work the application of experimental design, phenomenological modeling and artificial neural networks (ANN) in understanding chitosanases adsorption on ion exchange resin Streamline? DEAE have been investigated. The strain Paenibacillus ehimensis NRRL B-23118 was used for chitosanase production. EBA experiments were carried out using a column of 2.6 cm inner diameter with 30.0 cm in height that was coupled to a peristaltic pump. At the bottom of the column there was a distributor of glass beads having a height of 3.0 cm. Assays for residence time distribution (RTD) revelead a high degree of mixing, however, the Richardson-Zaki coefficients showed that the column was on the threshold of stability. Isotherm models fitted the adsorption equilibrium data in the presence of lyotropic salts. The results of experiment design indicated that the ionic strength and superficial velocity are important to the recovery and purity of chitosanases. The molecular mass of the two chitosanases were approximately 23 kDa and 52 kDa as estimated by SDS-PAGE. The phenomenological modeling was aimed to describe the operations in batch and column chromatography. The simulations were performed in Microsoft Visual Studio. The kinetic rate constant model set to kinetic curves efficiently under conditions of initial enzyme activity 0.232, 0.142 e 0.079 UA/mL. The simulated breakthrough curves showed some differences with experimental data, especially regarding the slope. Sensitivity tests of the model on the surface velocity, axial dispersion and initial concentration showed agreement with the literature. The neural network was constructed in MATLAB and Neural Network Toolbox. The cross-validation was used to improve the ability of generalization. The parameters of ANN were improved to obtain the settings 6-6 (enzyme activity) and 9-6 (total protein), as well as tansig transfer function and Levenberg-Marquardt training algorithm. The neural Carlos Eduardo de Ara?jo Padilha dezembro/2013 9 networks simulations, including all the steps of cycle, showed good agreement with experimental data, with a correlation coefficient of approximately 0.974. The effects of input variables on profiles of the stages of loading, washing and elution were consistent with the literature / A adsor??o em leito expandido (ALE) ? uma t?cnica integrativa que alia conceitos de cromatografia e fluidiza??o de s?lidos. A diversidade de par?metros envolvidos e seus efeitos sinerg?ticos dificultam a tarefa de otimiza??o da opera??o. Felizmente, algumas ferramentas matem?ticas foram desenvolvidas de modo a direcionar as investiga??es do sistema ALE. Assim, o presente trabalho prop?e a aplica??o do planejamento experimental, modelagem fenomenol?gica e redes neurais artificiais (RNAs) na compreens?o da adsor??o de quitosanases na resina de troca i?nica Streamline? DEAE. A cepa Paenibacillus ehimensis NRRL B-23118 foi respons?vel pela produ??o das quitosanases. Nos ensaios de adsor??o usando o leito na forma expandida foi utilizada uma coluna de 2,6 cm de di?metro por 30,0 cm de altura, acoplada a uma bomba perist?ltica. Na base da coluna existia um distribuidor de microesferas de vidro com altura de 3,0 cm. Os ensaios de determina??o de tempo de resid?ncia (DTR) revelaram elevado grau de mistura, entretanto, os coeficientes de Richardson-Zaki mostraram que a coluna estava no limiar da estabilidade. Pelas regress?es das isotermas puderam-se ajustar os dados de equil?brio de adsor??o, na presen?a de diferentes sais da escala liotr?pica. O resultado do planejamento apontou que a for?a i?nica e a velocidade influenciam a recupera??o e pureza das quitosanases. As massas moleculares das duas esp?cies de quitosanases foram estimadas por SDS-PAGE, obtendo-se aproximadamente 23 kDa e 52 kDa. A modelagem fenomenol?gica foi direcionada para descrever as opera??es em batelada e na coluna cromatogr?fica. As simula??es foram executadas no Microsoft Visual Studio, usando a linguagem Fortran. O modelo de taxa constante ajustou-se ?s curvas cin?ticas com excel?ncia, nas condi??es de atividade iniciais 0,232, 0,142 e 0,079 UA/mL. As curvas de ruptura simuladas apresentaram algumas disparidades com os dados experimentais, principalmente quanto ? inclina??o. Os testes de sensibilidade do modelo sobre a velocidade superficial, dispers?o axial e concentra??o inicial mostraram conformidade com artigos publicados. A rede neural foi constru?da no ambiente MATLAB, por meio da Neural Network Toolbox. A valida??o cruzada foi usada para melhorar a capacidade de generaliza??o. Carlos Eduardo de Ara?jo Padilha dezembro/2013 6 Aperfei?oaram-se os par?metros da RNA at? se obter as configura??es 6-6 (atividade enzim?tica) e 9-6 (prote?nas totais), fun??o de ativa??o tansig e algoritmo de treinamento Levenberg-Marquardt. As simula??es da rede neural, incluindo todo o ciclo da opera??o, mostraram boa concord?ncia com os dados experimentais, com coeficiente de correla??o da ordem de 0,974. Os efeitos das vari?veis de entrada sobre os perfis das etapas de carga, lavagem e elui??o foram compat?veis com a literatura