Predicting the secondary structure of proteins is important in biochemistry because the 3D structure can be determined from the local folds that are found in secondary structures. Moreover, knowing the tertiary structure of proteins can assist in determining their functions. The objective of this thesis is to compare the performance of Neural Networks (NN) and Support Vector Machines (SVM) in predicting the secondary structure of 62 globular proteins from their primary sequence. For each NN and SVM, we created six binary classifiers to distinguish between the classes’ helices (H) strand (E), and coil (C). For NN we use Resilient Backpropagation training with and without early stopping. We use NN with either no hidden layer or with one hidden layer with 1,2,...,40 hidden neurons. For SVM we use a Gaussian kernel with parameter fixed at = 0.1 and varying cost parameters C in the range [0.1,5]. 10- fold cross-validation is used to obtain overall estimates for the probability of making a correct prediction. Our experiments indicate for NN and SVM that the different binary classifiers have varying accuracies: from 69% correct predictions for coils vs. non-coil up to 80% correct predictions for stand vs. non-strand. It is further demonstrated that NN with no hidden layer or not more than 2 hidden neurons in the hidden layer are sufficient for better predictions. For SVM we show that the estimated accuracies do not depend on the value of the cost parameter. As a major result, we will demonstrate that the accuracy estimates of NN and SVM binary classifiers cannot distinguish. This contradicts a modern belief in bioinformatics that SVM outperforms other predictors.

Orientador: Flávia Cristina Rodrigues Lisoni / Coorientador: Sonia Maria Oliani / Banca: Érika Cristina Pavarino / Banca: Claúdia Marcia Aparecida Carareto / Resumo: INTRODUÇÃO: A inflamação é caracterizada pelo acúmulo de leucócitos nos tecidos oculares, podendo afetar qualquer parte do olho, sendo um processo doloroso, associado à fotofobia e podendo resultar em complicações secundárias. Em geral, no tratamento dessas inflamações intra-oculares, os glicocorticóides são os medicamentos freqüentemente administrados. Porém, devido aos efeitos adversos, existe uma demanda óbvia para o desenvolvimento de novas estratégias terapêuticas. Entre elas, a proteína anexina A1 (ANXA1) tem mostrado a sua participação ativa nos processos inflamatórios. OBJETIVO: Em função desses dados, foi realizado o presente trabalho que teve como objetivo geral avaliar a influência desse tratamento anti-inflamatório nas células epiteliais pigmentadas da retina humana (ARPE-19), sobre a morfologia, proliferação, migração e na expressão gênica e proteica. MATERIAL E MÉTODOS: As células ARPE-19 foram cultivadas em meio completo e estimuladas pelo Lipopolissacarideo bacteriano (LPS), e em outro experimento foi acrescentado o tratamento com o peptídeo ANXA1 Ac2-26 , para avaliar o possível efeito anti-inflamatório dessa proteína, nos tempos de 1, 2, 4, 24, 48 e 72 horas sobre a morfologia, proliferação e migração celular. Após análises estatísticas, que evidenciaram o tempo de 72 horas, foi realizado um novo cultivo celular para a extração do RNA e conduzida a técnica de Hibridização Subtrativa Rápida (RaSH). A validação de cinco genes encontrados como diferencialmente expressos foi realizada por RT-qPCR. Além dessas técnicas, também foi analisada a expressão das citocinas anti e pró-inflamatórias pelo ensaio Multiplex. RESULTADOS: O ANXA1Ac2-26 não modificou a morfologia das células ARPE-19, entretanto diminui a proliferação e aumenta a migração celular. Pela tecnologia de... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: INTRODUCTION: The inflammation is characterized by the accumulation of leukocytes in ocular tissues, and it can affect any part of the eye, been a painful process, associated with photophobia and may result in secondary complications. In general, for the treatment of these intraocular inflammations, the glucocorticoids are the drugs frequently administered. However, due to its side effects, there is an obvious demand for developing new therapeutic strategies. Among them, the protein annexin A1 (ANXA1) has shown its active participation in inflammatory processes. AIM: In the present study we investigated the influence of anti-inflammatory treatment in human retinal pigment epithelial cells (ARPE-19), on the morphology, proliferation, migration and gene and protein expression. MATERIALS AND METHODS: The ARPE-19 cells were cultured in complete medium and stimulated by bacterial lipopolysaccharide (LPS), and in another experiment was added treatment with ANXA1Ac2-26 peptide, to evaluate the possible anti- inflammatory effects of this protein, in 1, 2, 4, 24, 48 and 72 hours on the morphology, cell proliferation and migration. After statistical analyzes, which demonstrated that the time 72 hours is the best, was performed a new culture bottles for RNA extraction technique and conducted the Rapid Subtractive Hybridization (RaSH). The validation of five genes founded differentially expressed was done by RT-qPCR. In addition to these techniques was also analyzed the expression of anti and pro-inflammatory cytokines by Multiplex assay. RESULTS: The ANXA1Ac2-26 did not alter the morphology of the ARPE- 19 cells, however decreased proliferation and increased cell migration. After statistical analysis, it was evident that the time of 72 hours was the best to proceed with the methodology. By RaSH technology... (Complete abstract click electronic access below) / Mestre

Foi desenvolvido um conjunto de peças plásticas que permitem a montagem e representação dos aminoácidos mais comuns, bem como a construção de estruturas protéicas. Durante e após o desenvolvimento o material foi submetido a várias etapas de avaliação por professores (do ensino básico e universitário), alunos de pós-graduação e de graduação. A primeira etapa foi o desenvolvimento dos modelos em ambiente computacional, seguida da prototipagem das peças. Após discussão com a comunidade científica (apresentados na XXXVI Reunião Anual da SBBq em 2007) as sugestões foram implementadas nos modelos computacionais. Quatro moldes para injeção termoplástica foram projetados, detalhados e construídos, sob nossa orientação. As peças representando as estruturas que compõe os aminoácidos e ligações foram produzidas em grande escala e iniciou-se o processo final de avaliação. As peças apresentaram boas relações geométricas com as fórmulas estruturais dos aminoácidos obtidas de bancos de dados e livros didáticos. As conexões C&alpha;-amina e C&alpha;-carboxila permitem verificar a liberdade de rotação característica das cadeias polipeptídicas e as possibilidades dos ângulos de torção &Psi; e &Phi;, visualizando a restrição de rotação da ligação peptídica. Montando um conjunto de aminoácidos é possível construir uma cadeia polipeptídica e, através das ligações de hidrogênio, montar as estruturas secundárias principais (hélice-&alpha; e estruturas &beta;). Duas avaliações preliminares foram realizadas e a avaliação final ocorreu em uma oficina de atividades com 256 professores das áreas de ciências da natureza da rede publica do Estado de SP. Os resultados da avaliação foram extremamente positivos, sendo importante destacar a quantidade e o teor dos comentários elogiosos ao potencial de utilização do material, notadamente, dos professores de biologia e química. O material poderá inclusive auxiliar no preenchimento de lacunas conceituais que existem na formação dos professores e que foram observadas durante as atividades de avaliação. Este conjunto de peças, organizado na forma de um kit: Construindo Estruturas de Aminoácidos e Proteínas, foi submetido à avaliação do MEC e certificado por este órgão, passando a integrar o Guia de Tecnologias Educacionais 2008.<b/> / It was developed a set of plastic pieces that allow the assembly and representation of the most common amino acids, as well as the construction of protein structures. During and after development the material was submitted to several stages of evaluation by teachers (primary and university), graduate and undergraduate students. The first step was the development of models in the computing environment, followed by prototyping of parts. After discussion with the scientific community (presented at the XXXVI Annual Meeting of SBBq in 2007) suggestions were implemented in the computational models. Four thermoplastic injection molds were designed, detailed and constructed under our supervision. Parts representing the structures of amino acids and bonds were produced in large scale and it was started the final process of evaluation. The pieces had good geometric relationships with the structural formulas of amino acids obtained from databases and textbooks. The connections C&alpha;-amine and C&alpha;-arboxyl permit to check the freedom of rotation of the polypeptide chains and the possibility of torsion angles &Phi; and &Psi;, visualizing the restriction of rotation of the peptide bond. Assembling a set of amino acids is possible to build a polypeptide chain and, through hydrogen bonding, to assemble the main secondary structures (&alpha;-helix and &beta;-structures). Two preliminary evaluations were conducted and the final evaluation took place in a workshop with 256 teachers of the fields of natural sciences from public schools of the São Paulo State. The results of the evaluation were extremely positive and it is important to highlight the amount and content of approving comments for the potential of use of the material, especially from biology and chemistry teachers. The material may even assist in filling in conceptual gaps that exist in teacher instruction and that were observed during the evaluation activities. This set of pieces, arranged in the form of a kit: Building Structures of Amino Acids and Proteins, was submitted to MEC and certified by this organization, starting to integrate the Guide of Educational Technology 2008<b/>.

Orientadores: Beatriz Gomes Guimarães, Nilson Ivo Tonin Zanchin / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Biologia / Made available in DSpace on 2018-08-16T19:49:15Z (GMT). No. of bitstreams: 1 Giuseppe_PriscilaOliveirade_D.pdf: 13819393 bytes, checksum: 088685c4fb029e25a7f5c7b7f6e218cf (MD5) Previous issue date: 2010 / Resumo: Leptospira interrogans é uma bactéria espiroqueta que causa a leptospirose, uma zoonose de distribuição mundial que afeta mais de 500.000 pessoas anualmente. Pouco se sabe sobre a biologia de leptospiras, o que dificulta a elaboração de novas estratégias de prevenção e de tratamento contra a doença. Cerca de 60 % dos genes de L. interrogans codifica proteínas que não apresentam similaridade de sequência significativa com proteínas de função conhecida. Como a estrutura cristalográfica de uma proteína pode revelar vários indícios funcionais, este trabalho visou à determinação da estrutura cristalográfica das proteínas LIC10793, LIC12922 e LIC10494 de L. interrogans, que são potencialmente localizadas no envelope celular e não são funcionalmente caracterizadas. A estrutura do antígeno LIC10793 (Lp49) foi resolvida a 2 Å de resolução e revelou que essa provável lipoproteína apresenta dois domínios. O domínio N-terminal de Lp49 possui um enovelamento do tipo Imunoglobulina e sua topologia diverge das formas padrão do motivo estrutural "chave grega" (Greek key motif). O domínio C-terminal consiste em um ?- propeller formado por sete folhas ?. Comparações locais não identificaram nenhum sítio catalítico conhecido em Lp49, mas análises de sua superfície revelaram a presença deprováveis sítios de ligação a proteínas. Com base nesses indícios, na provável localização de Lp49 na membrana externa e em sua antigenicidade, postula-se que Lp49 tenha uma função de interação com outras proteínas podendo desempenhar um papel na interação entre leptospiras e seus hospedeiros. A estrutura cristalográfica de LIC12922, determinada a 3,1 Å de resolução, revelou a presença de dois domínios. O domínio NC é estruturalmente relacionado a domínios que apresentam atividade chaperona, encontrados nas proteínas SurA e trigger factor de E. coli. O domínio parvulina de LIC12922 não apresenta atividade de peptidil prolil isomerase, mas possui um provável sítio de interação a proteína que inclui o sítio de reconhecimento ao substrato proposto para o domínio parvulina P1 de SurA. Análises filogenéticas sugerem que LIC12922 e as chaperonas extracitoplasmáticas SurA, PpiD e PrsA apresentam um ancestral comum. Com base nesses indícios e na provável localização de LIC12922 no periplasma, propõe-se que LIC12922 seja uma chaperona periplasmática envolvida na biogênese de proteínas da membrana externa. LIC10494 foi expressa, purificada e cristalizada. Refinamentos das condições de cristalização não foram suficientes para se obter cristais adequados ao experimento de difração. Análises da sua sequência evidenciaram que LIC10494 apresenta uma extensa região central intrinsecamente desordenada rica em resíduos de treonina. Assim como proteínas que possuem domínios intrinsecamente desestruturados, LIC10494 apresenta mobilidade mais lenta do que o esperado em SDS-PAGE, um volume de eluição menor do que o esperado em ensaios de gel filtração e uma considerável contribuição de configurações randômicas em seu espectro de dicroísmo circular / Abstract: Leptospira interrogans is a spirochaetal bacterium which causes leptospirosis, a worldwide spread zoonosis that affects more than 500,000 people annually. Little is known about the biology of leptospires, which difficults the development of new preventive and treatment strategies for the disease. About 60 % of the genes from L. interrogans encode for proteins that did not show significative sequence similarity with proteins of known function. Since the tridimensional structure of a protein can contribute to the understanding of its function, this work aimed at the crystallographic structure determination of the proteins LIC10793, LIC12922 and LIC10494 from L. interrogans, which are potentially situated at the cell envelope and are not functionally characterized. The structure of the antigen LIC10793 (Lp49) was determined at 2 Å resolution and revealed that this probable lipoprotein possesses two domains. The Lp49 N-terminal domain presents an Immunoglobulin-like fold and its topology diverges from the standard patterns of the Greek key motif. The C-terminal domain is a 7-bladed ?-propeller. Local structural comparisons did not identify known catalytic sites at Lp49, but surface analyses evidenced potential protein binding sites. Based on these results, the putative localization of Lp49 at the outer membrane and its role as an antigen, we postulate that Lp49 has a protein binding function involved in Leptospira-host interaction. The LIC12922 crystal structure, determined at 3.1 Å resolution, revealed two domains. The NC domain is structurally related to the chaperone domains of E. coli SurA and trigger factor proteins. The LIC12922 parvulin domain is devoid of peptidyl prolyl isomerase activity, but presents a putative protein binding site which includes the substrate recognition site proposed to the first parvulin domain of SurA. Phylogenetic analyses suggest that LIC12922 and the extracytoplasmic chaperones SurA, PpiD and PrsA have a common ancestor. Based on the structural and phylogenetic analyses and taking into account its probable periplasmic localization we postulate that LIC12922 is a periplasmic chaperone involved at the biogenesis of outer membrane proteins. The protein LIC10494 was expressed, purified and crystallized. In spite of extensive refinement of crystallization conditions the crystals were not adequate for diffraction experiments. Sequence analyses evidenced that LIC10494 has an extensive central region intrinsically disordered which is rich in threonine residues. Similarly to proteins that possess intrinsically disordered domains, LIC10494 presents mobility slower than expected at SDSPAGE, elution volume smaller than expected in gel filtration assays and a considerable contribution of random coil structures in circular dichroism spectrum / Doutorado / Genetica de Microorganismos / Doutor em Genetica e Biologia Molecular

Orientador: Carlos Henrique Inacio Ramos / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Biologia / Made available in DSpace on 2018-08-08T11:17:50Z (GMT). No. of bitstreams: 1 Cepeda_AnaOlivaTiroli_D.pdf: 3942958 bytes, checksum: 940e4a499333daa56fd9a4fc5899919c (MD5) Previous issue date: 2007 / Resumo: As proteínas estão envolvidas com as mais diversas funções biológicas. No entanto, para realizar sua função adequadamente, uma proteína deve estar enovelada, ou seja, em sua conformação nativa. Para garantir isso, existe nas células, um elaborado sistema que envolve chaperones moleculares, capaz de auxiliar na prevenção do enovelamento incorreto e da agregação de proteínas Chaperones, de uma maneira geral, são proteínas que ligam e estabilizam polipeptídeos, facilitando seu enovelamento correto sem contribuir com informações conformacionais. O aumento no número de doenças provocadas pelo enovelamento incorreto de proteínas que se depositam nos tecidos na forma de amilóides (também chamadas de doenças conformacionais), tem chamado a atenção para estudos de agregados protéicos, que outrora foram considerados artefatos quando se trabalhava com esse tipo de macromolécula. Nesse sentido, o estudo de chaperones tem ganhado um interesse particular, já que são fortes candidatos ao combate de doenças amiloloidogênicas. Neste trabalho, são apresentados estudos sobre duas famílias de chaperones, a Hsp40 da subfamília A humana e duas sHsps de classe I de cana-de-açúcar, as quais estão envolvidas com o reconhecimento e a apresentação de substratos (proteínas parcialmente desenoveladas) para outras famílias de chaperones responsáveis pelo processo de reenovelamento. Essas duas famílias de chaperones em particular são também conhecidas como 'holdases¿, e são muito diversas, característica necessária para interagir com a grande diversidade de substratos em potencial que existe na célula. As duas sHsps estudadas aqui, as mais expressas em cana-de-açúcar, e a caracterização de suas estruturas e suas eficiências como chaperones, tornou possível a elaboração de uma hipótese sobre o mecanismo de ação dessas proteínas em função do aumento de temperatura. Nesse sentido, é mostrado neste trabalho que sHsps, respondem ao aumento de temperatura passando por expansão conformacional, provavelmente para aumentar a superfície hidrofóbica para a interação com os substratos. O efeito do calor sobre a Hsp40 também foi estudado e os resultados mostraram que essa proteína forma agregados com propriedades amiloidogênicas. Esta é a primeira vez que tais características são descritas para um chaperone de eucarioto. De maneira geral, as implicações dos resultados apresentados aqui podem aumentar o conhecimento geral sobre chaperones e sobre a pesquisa de tratamentos para as doenças conformacionais / Abstract: Proteins are involved with a large variety of biological functions. However, to function properly, proteins must be folded, i.e., they must reach their native conformation. According to that, an elaborated system involving molecular chaperones exists in the cell that helps to prevent the incorrect folding of proteins and also their aggregation. Chaperones, in a general way, are proteins that bind and stabilize polypeptides, facilitating its correct folding without contributing with conformational information. The increasing number of diseases caused by the incorrect folding of proteins that deposit in the form of amyloids (also called conformational diseases) has raised the interest in the study of protein aggregates, which, not long ago, where considered just purification artifacts. In this way, the study of chaperones has gained particular interest because they are potential candidates against amyloidogenic diseases. In this work, we present studies on two families of chaperones, a human Hsp40 from subfamily A and two sugar cane sHsps from class I, which are involved in substrate (partially unfolded proteins) recognition and presentation to other chaperone families that are more active in the protein refolding process. These particular chaperones are also know as 'holdases¿ and they are usually diverse, a characteristic necessary to interact with a large variety of substrate in the cell. The two sHsps studied here are the most expressed in sugar cane and their structure and chaperone efficiency characterization made possible to elaborate a hypothesis on the mechanism of action of these proteins when temperature increases. In that matter, we were able to show that sHsps respond to an increase in temperature by undergoing conformational expansion, likely to increase the hydrophobic area for substrate interaction. The effect of heat on Hsp40 has also been studied and our results showed that this protein form aggregates with amyloidogenic properties. To our knowledge, this is the first time that such characteristics are described for an eukaryotic chaperone. To sum up, we believe that the implications of the results shown here may add to the general knowledge on chaperones and to the search of a treatment for conformational diseases / Doutorado / Bioquimica / Doutor em Biologia Funcional e Molecular

The function of a biological molecule is linked to its underlying structure, and determination of that structure can lead to significant insights into its function and how this is performed. There already exist a number of important tools in structural biology, however, the pulsed electron paramagnetic resonance (EPR) technique called pulsed electron-electron double resonance (PELDOR) is the only one capable of accurately measuring isolated distances between attached spin-labels over the range of ~2 to 10 nm, a range which is usually impossible to measure directly with other techniques such as nuclear magnetic resonance (NMR) and X-ray crystallography. This can provide constraints for refinement of structures determined from NMR and X-ray crystallography, or insights into protein docking and protein mechanics. With recent developments in EPR spectrometer instrumentation and spin-labelling it has become possible to conduct PELDOR experiments in the high field EPR regime ( > 3 Tesla) where measurement sensitivity is increased. These experiments can reveal relative orientations of nitroxide spin-labels in complement to their separation, however, analysis and interpretation of these results has been difficult to perform routinely. This thesis presents a characterisation of the high field spectrometer HiPER showing that it is well suited when optimised for making PELDOR experiments. To perform analysis of PELDOR signals from this spectrometer custom signal simulation code has been written. Two case studies are presented. The first relates to the use of the Rx spin label with the PELDOR experiment to derive orientation information from the spin labelled protein Vps75. The recently developed spin label Rx is proposed to attach more rigidly to underlying structure, offering potentially increased accuracy in determination of structure constraints and additional information about relative orientations of different structural features. An orientation selective PELDOR study is presented which compares molecular dynamics (MD) simulations of spin labels attached to sites on the α-helix of the protein Vps75. This has shown great potential for utilising the Rx spin label in a repeatable way on α-helix residue sites for determination of structural constraints. The second case relates to orientation selective PELDOR measurements of spin labelled oligomeric membrane protein structures. High field PELDOR offers great potential in increasing measurement sensitivity and accuracy of structural constraints in oligomeric proteins. A methodology of signal analysis for this class of protein is presented along with measurements of the membrane channel protein MscS. Difficulties of PELDOR measurement on these labelled proteins are discussed and observed relaxation of the spin echo, relevant to pulsed EPR experiments, are investigated and possible mechanisms are presented.

Protein folding, translocation, oligomeric rearrangement and degradation are vital functions to obtain correctly folded proteins in any cell. The constitutive or stress-induced members of each of the heat shock protein (Hsp) families, namely Hsp70 and Hsp40, make up the Hsp70/Hsp40 chaperone system. The Hsp40 J-domain is important for the Hsp70-Hsp40 interaction and hence function. The type-II Hsp40 proteins, Homo sapiens DnaJ 1a (Hsj1a) and Plasmodium falciparum DnaJ 4 (Pfj4), are structurally similar suggesting possible similar roles during malarial infection. This thesis has focussed on identifying whether Hsj1a and Pfj4 are functionally similar in their interaction with potential partner Hsp70 chaperones. Analysis in silico also showed Pfj4 to have a potential chaperone domain, a region resembling a ubiquitin-interacting motif (UIM) corresponding to UIM1 of HsjIa, and another highly conserved region was noted between residues 232-241. The highly conserved regions within the Hsp40 J-domains, and those amino acids therein, are suggested to be responsible for mediating this Hsp70-Hsp40 partner interaction. The thermosensitive dnaJ cbpA Escherichia coli OD259 mutant strain producing type-I Agrobacterium tumefaciens DnaJ (AgtDnaJ) was used as a model heterologous expression system in this study. AgtDnaJ was able to replace the lack of two E coli Hsp40s in vivo, DnaJ and CbpA, whereas AgtDnaJ(H33Q) was unable to. AgtDnaJ-based chimeras containing the swapped J-domains of similar type-II Hsp40 proteins, namely Hsj1Agt and Pfj4Agt, were also able to replace these in E. coli OD259. Conserved J-domain amino acids were identified and were substituted in these chimeras. Of these mutant proteins, Hsj IAgt(L8A), Hsj1Agt(R24A), Hsj1Agt(H31Q), Pfj4Agt(L 11A) and Pfj4Agt(H34Q) were not able to replace the E. coli Hsp40s, whilst Pfj4Agt(Y8A) and Pfj4Agt(R27A) were only able to partially replace them. This shows the leucine of helix I and the histidine of the loop region are key in the in vivo function of both proteins and that the arginine of helix II is key for Hsj1a. The histidine-tagged Hsj1a protein was also successfully purified from the heterologous system. The in vitro stimulated ATPase activity of human Hsp70 by Hsj1a was found to be approximately 14 nmol Pí[subscript]/min/mg, and yet not stimulated by Pfj4, suggesting a possible species-specific interaction is occurring.

The 70 kDa heat shock proteins (Hsp70s) are vital for normal protein folding, as they stabilise the unfolded state of nascent polypeptides, allowing these sufficient time to attain a correct tertiary structure. Hsp70s are aided by the DnaJ-like family of proteins, which interact with Hsp70s in order to enhance the chaperone activity of these proteins. DnaJ-like proteins contain a J domain, a seventy amino acid domain consisting of four α-helices, which is defined by the presence of an invariant tripeptide of histidine, proline and aspartic acid (HPD motif). This motif is key to the interaction between DnaJ-like proteins and Hsp70s. This thesis has focused on determining the presence of other conserved residues in the J domain and their role in mediating the interaction of DnaJ-like proteins with partner Hsp70s. DnaJ-like proteins from Agrobacterium tumefaciens RUOR were isolated and used as a model system. A. tumefaciens DnaJ (Agt DnaJ) was able to replace the lack of E. coli DnaJ in an E. coli null mutant strain, however, additional A. tumefaciens DnaJ-like proteins Agt DjC1/DjlA, Agt DjC2 and Agt DjC5 were unable to complement for the lack of E. coli DnaJ. Replacement of the Agt DnaJ J domain with J domains from these proteins resulted in non-functional chimeric proteins, despite some sequence conservation. The kinetics of the basal specific ATPase activity of Agt DnaK, and its ability to have this activity stimulated by Agt DnaJ and Agt DnaJ-H33Q were also investigated. Stimulation of the ATPase activity by Agt DnaJ ranged between 1.5 to 2 fold, but Agt DnaJ-H33Q was unable to stimulate the basal ATPase activity. Conserved amino acids in the J domain were identified in silico, and these residues were substituted in the J domain of Agt DnaJ. The ability of these derivative proteins to replace E. coli DnaJ was investigated. Alterations in the HPD motif gave rise to proteins unable to complement for lack of E. coli DnaJ, consistent with literature. Agt DnaJ-R26A was unable to replace E. coli DnaJ suggesting that Arg26 could be key to the interaction with partner Hsp70s. Agt DnaJ-D59A was unable to replace E. coli DnaJ; substitutions in Asp59 have not previously been shown to impact on the function of DnaJ. Substituting Arg63 in Agt DnaJ abrogated the levels of complementation. Substitution of several structural residues was also found to disrupt the in vivo function of Agt DnaJ suggesting that the maintenance of the structural integrity of the J domain was important for function. This study has identified a number of residues critical to the structure and function of the J domain of Agt DnaJ, and potentially of general importance as molecular determinants for DnaJ-Hsp70 interaction.

Plasmodium falciparum is the pathogen that is responsible for the most virulent, severe and dangerous form of human malaria infection, accounting for nearly a million deaths every year. To survive and develop in the unusual environment of the red blood cells, the parasite causes structural remodelling of the host cell and biochemical changes through the export of virulence factors. Among the exportome are the molecular chaperones of the heat shock protein family, of which Hsp40s and Hsp70s are prominent. PF A0660w, a type II P. falciparum Hsp40, has been shown to be exported in complex with PfHsp70-x into the infected erythrocyte, suggesting possible functional interactions. However, the chaperone properties of PF A0660w and its interactions with proteins of parasite and human origin are yet to be investigated. Using a codon optimised coding region, PF A0660w was successfully expressed in E. coli M 15 [pREP4] cells. However, the expressed protein was largely deposited as insoluble pellet, and analysis of the pellets revealed a high percentage of PF A0660w, characteristic of inclusion body formation. PF A0660w was purified from inclusion bodies using additive enhanced solubilisation and refolding buffers followed by nickel affinity chromatography. SDS-PAGE and western analysis revealed that the purified protein was of high purity. Size exclusion chromatography showed that the protein existed as a monomer in solution and the secondary structure analysis using Fourier transformed infrared spectroscopy (FTIR) confirmed the success of the refolding approach. Its monomeric state suggests that PF A0660w may be functionally different from other Hsp40 that form dimers and that for PF A0660w, dimer formation may not be needed to maintain the stability of the protein in solution, but may occur in response to functional necessities during its interaction with partner Hsp70. PFA0660w was able to significantly stimulate the ATPase activity ofPfl-Isp70-x but not Pfl-Isp70-1 or human Hsp70 (HsHsp70), suggesting a specific functional interaction. Also, PF A0660w produced a dose dependent suppression of rhodanese aggregation and cooperated with Pfl-Isp70-1, PfHsp70-x and HsHsp70 to cause enhanced aggregation suppression. Its ability to independently suppress aggregation may help to maintain substrates in an unfolded conformation for eventual transfer to partner Hsp70s during refolding processes. Also, the in vivo characterisation using a PF A0660w peptide specific antibody confirmed that PF A0660w was exported into the cytosol of infected erythrocytes. Its lack of induction upon heat shock suggests that PF A0660w may not be involved in the response of the parasite to heat stress. Overall, this study has provided the first heterologous over-expression, purification and biochemical evidence for the possible functional role of PF A0660w, and has thereby provided the needed background for further exploration of this protein as a potential target for drug discovery.

Orientadores: Siome Klein Goldenstein, Carlile Campos Lavor / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Computação / Made available in DSpace on 2018-08-22T06:27:38Z (GMT). No. of bitstreams: 1 Sendin_IvandaSilva_D.pdf: 14139008 bytes, checksum: d64497ed260da601f2dde025683df0d0 (MD5) Previous issue date: 2012 / Resumo: A determinação da estrutura de uma proteína usando dados de Ressonância Magnética Nuclear precisa lidar com incertezas provenientes do experimento laboratorial. Neste trabalho, apresentamos um método híbrido utilizando aritmética afim e propagação de incerteza por partículas para o tratamento e o controle de incertezas. Aplicado no cálculo da estrutura de proteínas, o método proposto é capaz de gerar estruturas de proteínas que atendem satisfatoriamente as restrições do problema / Abstract: The protein structure determination using Nuclear Magnetic Resonance data uses imprecise information from laboratorial experiments. In this work we introduce a new hybrid method that combines affine arithmetic and particles to uncertainty propagation and control. Applied to protein structure determination this new method was able to determine protein structures that satisfy most of problem constraints / Doutorado / Ciência da Computação / Doutor em Ciência da Computação