Searching for novel gene functions in yeast : identification of thousands of novel molecular interactions by protein-fragment complementation assay followed by automated gene function prediction and high-throughput lipidomics

La compréhension de processus biologiques complexes requiert des approches expérimentales et informatiques sophistiquées. Les récents progrès dans le domaine des stratégies génomiques fonctionnelles mettent dorénavant à notre disposition de puissants outils de collecte de données sur l’interconnectivité des gènes, des protéines et des petites molécules, dans le but d’étudier les principes organisationnels de leurs réseaux cellulaires. L’intégration de ces connaissances au sein d’un cadre de référence en biologie systémique permettrait la prédiction de nouvelles fonctions de gènes qui demeurent non caractérisées à ce jour. Afin de réaliser de telles prédictions à l’échelle génomique chez la levure Saccharomyces cerevisiae, nous avons développé une stratégie innovatrice qui combine le criblage interactomique à haut débit des interactions protéines-protéines, la prédiction de la fonction des gènes in silico ainsi que la validation de ces prédictions avec la lipidomique à haut débit. D’abord, nous avons exécuté un dépistage à grande échelle des interactions protéines-protéines à l’aide de la complémentation de fragments protéiques. Cette méthode a permis de déceler des interactions in vivo entre les protéines exprimées par leurs promoteurs naturels. De plus, aucun biais lié aux interactions des membranes n’a pu être mis en évidence avec cette méthode, comparativement aux autres techniques existantes qui décèlent les interactions protéines-protéines. Conséquemment, nous avons découvert plusieurs nouvelles interactions et nous avons augmenté la couverture d’un interactome d’homéostasie lipidique dont la compréhension demeure encore incomplète à ce jour. Par la suite, nous avons appliqué un algorithme d’apprentissage afin d’identifier huit gènes non caractérisés ayant un rôle potentiel dans le métabolisme des lipides. Finalement, nous avons étudié si ces gènes et un groupe de régulateurs transcriptionnels distincts, non préalablement impliqués avec les lipides, avaient un rôle dans l’homéostasie des lipides. Dans ce but, nous avons analysé les lipidomes des délétions mutantes de gènes sélectionnés. Afin d’examiner une grande quantité de souches, nous avons développé une plateforme à haut débit pour le criblage lipidomique à contenu élevé des bibliothèques de levures mutantes. Cette plateforme consiste en la spectrométrie de masse à haute resolution Orbitrap et en un cadre de traitement des données dédié et supportant le phénotypage des lipides de centaines de mutations de Saccharomyces cerevisiae. Les méthodes expérimentales en lipidomiques ont confirmé les prédictions fonctionnelles en démontrant certaines différences au sein des phénotypes métaboliques lipidiques des délétions mutantes ayant une absence des gènes YBR141C et YJR015W, connus pour leur implication dans le métabolisme des lipides. Une altération du phénotype lipidique a également été observé pour une délétion mutante du facteur de transcription KAR4 qui n’avait pas été auparavant lié au métabolisme lipidique. Tous ces résultats démontrent qu’un processus qui intègre l’acquisition de nouvelles interactions moléculaires, la prédiction informatique des fonctions des gènes et une plateforme lipidomique innovatrice à haut débit , constitue un ajout important aux méthodologies existantes en biologie systémique. Les développements en méthodologies génomiques fonctionnelles et en technologies lipidomiques fournissent donc de nouveaux moyens pour étudier les réseaux biologiques des eucaryotes supérieurs, incluant les mammifères. Par conséquent, le stratégie présenté ici détient un potentiel d’application au sein d’organismes plus complexes.

Gene Ontology Function prediction in Mollicutes using Protein-Protein Association Networks

Abstract Background: Many complex systems can be represented and analysed as networks. The recent availability of large-scale datasets, has made it possible to elucidate some of the organisational principles and rules that govern their function, robustness and evolution. However, one of the main limitations in using protein-protein interactions for function prediction is the availability of interaction data, especially for Mollicutes. If we could harness predicted interactions, such as those from a Protein-Protein Association Networks (PPAN), combining several protein-protein network function-inference methods with semantic similarity calculations, the use of protein-protein interactions for functional inference in this species would become more potentially useful. Results: In this work we show that using PPAN data combined with other approximations, such as functional module detection, orthology exploitation methods and Gene Ontology (GO)-based information measures helps to predict protein function in Mycoplasma genitalium. Conclusions: To our knowledge, the proposed method is the first that combines functional module detection among species, exploiting an orthology procedure and using information theory-based GO semantic similarity in PPAN of the Mycoplasma species. The results of an evaluation show a higher recall than previously reported methods that focused on only one organism network.

RNAi-mediated silencing of vitellogenin gene function turns honeybee (Apis mellifera) workers into extremely precocious foragers

The switch from within-hive activities to foraging behavior is a major transition in the life cycle of a honeybee (Apis mellifera) worker. A prominent regulatory role in this switch has long been attributed to juvenile hormone (JH), but recent evidence also points to the yolk precursor protein vitellogenin as a major player in behavioral development. In the present study, we injected vitellogenin double-stranded RNA (dsVg) into newly emerged worker bees of Africanized genetic origin and introduced them together with controls into observation hives to record flight behavior. RNA interference-mediated silencing of vitellogenin gene function shifted the onset of long-duration flights (> 10 min) to earlier in life (by 3-4 days) when compared with sham and untreated control bees. In fact, dsVg bees were observed conducting such flights extremely precociously, when only 3 days old. Short-duration flights (< 10 min), which bees usually perform for orientation and cleaning, were not affected. Additionally, we found that the JH titer in dsVg bees collected after 7 days was not significantly different from the controls. The finding that depletion of the vitellogenin titer can drive young bees to become extremely precocious foragers could imply that vitellogenin is the primary switch signal. At this young age, downregulation of vitellogenin gene activity apparently had little effect on the JH titer. As this unexpected finding stands in contrast with previous results on the vitellogenin/JH interaction at a later age, when bees normally become foragers, we propose a three-step sequence in the constellation of physiological parameters underlying behavioral development.

Molecular Modeling Approaches for Determining Gene Function: application to a Putative Poly-A Binding Protein from Leishmania amazonensis (LaPABP)

The great expansion in the number of genome sequencing projects has revealed the importance of computational methods to speed up the characterization of unknown genes. These studies have been improved by the use of three dimensional information from the predicted proteins generated by molecular modeling techniques. In this work, we disclose the structure-function relationship of a gene product from Leishmania amazonensis by applying molecular modeling and bioinformatics techniques. The analyzed sequence encodes a 159 aminoacids polypeptide (estimated 18 kDa) and was denoted LaPABP for its high homology with poly-A binding proteins from trypanosomatids. The domain structure, clustering analysis and a three dimensional model of LaPABP, basically obtained by homology modeling on the structure of the human poly-A binding protein, are described. Based on the analysis of the electrostatic potential mapped on the model's surface and conservation of intramolecular contacts responsible for folding stabilization we hypothesize that this protein may have less avidity to RNA than it's L. major counterpart but still account for a significant functional activity in the parasite. The model obtained will help in the design of mutagenesis experiments aimed to elucidate the mechanism of gene expression in trypanosomatids and serve as a starting point for its exploration as a potential source of targets for a rational chemotherapy.

Targeted Gene Deletion and In Vivo Analysis of Putative Virulence Gene Function in the Pathogenic Dermatophyte Arthroderma benhamiae.

Dermatophytes cause the majority of superficial mycoses in humans and animals. However, little is known about the pathogenicity of this specialized group of filamentous fungi, for which molecular research has been limited thus far. During experimental infection of guinea pigs by the human pathogenic dermatophyte Arthroderma benhamiae, we recently detected the activation of the fungal gene encoding malate synthase AcuE, a key enzyme of the glyoxylate cycle. By the establishment of the first genetic system for A. benhamiae, specific ΔacuE mutants were constructed in a wild-type strain and, in addition, in a derivative in which we inactivated the nonhomologous end-joining pathway by deletion of the A. benhamiae KU70 gene. The absence of AbenKU70 resulted in an increased frequency of the targeted insertion of linear DNA by homologous recombination, without notably altering the monitored in vitro growth abilities of the fungus or its virulence in a guinea pig infection model. Phenotypic analyses of ΔacuE mutants and complemented strains depicted that malate synthase is required for the growth of A. benhamiae on lipids, major constituents of the skin. However, mutant analysis did not reveal a pathogenic role of the A. benhamiae enzyme in guinea pig dermatophytosis or during epidermal invasion of the fungus in an in vitro model of reconstituted human epidermis. The presented efficient system for targeted genetic manipulation in A. benhamiae, paired with the analyzed infection models, will advance the functional characterization of putative virulence determinants in medically important dermatophytes.

Characterization of PHO1 and PHO1;H1 gene function in Arabidopsis thaliana

Summary Phosphorus is one of the major macronutrients required for plant growth and development. Plant roots acquire phosphorus as inorganic phosphate (Pi), which is further distributed to the shoot, via the transpiration stream and root pressure, where Pi is imported again into cells. PHO1 in Arabidopsis has been identified as a protein involved in the loading of Pi into the root xylem. PHO1 does not have any homology to described Pi transporters including the Pht1 family of H+/ Pi cotransporters. PHO1 bears two domains, SPX and EXS domains, previously identified in Saccharomyces cerevisiae proteins involved in Pi transport and/or sensing, or in sorting proteins to endomembranes. Phylogenetic analysis of the PHO1 gene family revealed the presence of three clusters, with PHO1 and PHO1;H1 forming one cluster. The biological significance behind this cluster was demonstrated by the complementation of the pho1 mutant with only PHO1 and PHO1;H1, of all the PHO1 family members, when expressed under the PHO1 promoter. PHO1 has been shown to be expressed mostly in the root vascular cylinder and at low level in the shoot. PHO1;H1 had a different expression pattern, being expressed in both root and shoot vascular cylinder to the same level, with the levels in leaves increasing with the leaf maturity, suggesting additional role of PHO1;H1 in the Pi mobilization in leaves. In order to further explore the role of PHO1, Pi dynamics was studied on plants expressing PHO1 at different levels compared to the wild type: PHO1 overexpressors, PHO1 underexpressors and the pho1 mutant. Overexpression of the PHO1 protein in the shoot vascular tissue was shown to lead to increased Pi efflux out of the leaf cells and Pi accumulation in the shoot xylem apoplast compared to wild type, confirming the hypothesized role of PHO1 in xylem loading with Pi. The overexpression of PHO1 in the shoot was responsible far both changed Pi dynamic and stunted growth of PHO1 overexpressors, as shown by grafting experiments between wild type and PHO1 overexpressor. We found a ca. 2 fold decrease of shoot phosphorus and a 5-10 fold decrease in vacuolar Pi content in the PHO1 underexpressors and the pho1 null mutant compared to wild type, consistent with the role of PHO1 in the transfer of Pi from the root to the shoot. Shoot Pi deficiency results in a poor growth of the pho1 mutant. Grafting experiments between pho1 and wild type confirmed that both Pi deficiency and stunt growth of the pho1 mutant were dependent on the pho1 root, further supporting the importance of PHO1 in the root xylem loading with Pi. The pho1 mutant and the PHO1 underexpressors accumulated 8-15 fold more Pi in the root relative to wild type. In contrast to the pho1 mutant, the growth of PHO1 underexpressors was not impaired by the low shoat Pi content. This finding suggests that either PHO1 protein or root Pi concentration is important in Pi signaling and development of Pi deficiency symptoms leading to reduced growth. Résumé Le phosphore est l'un des nutriments essentiels à la croissance et au développement des plantes. Les racines absorbent le phosphore sous forme de phosphate inorganique (Pi) qui est dirigé, par la transpiration et la pression de la racine, vers les feuilles où le phosphate est acquis par les cellules. La protéine PHO1 a été démontrée indispensable au chargement du Pi dans le xylème des racines d'Arabidopsis. PHO1 ne démontre pas d'homologie aux transporteurs de Pi connus, incluant la famille Pht1 de cotransporteurs H+/Pi qui ont comme fonction le transport du phosphate à l'intérieur de la cellule. PHO1 contient deux domaines, SPX et EXS, aussi présents dans des protéines de Saccharomyces cerevisiae impliquées dans le transport ou la perception du phosphate, ou dans la localisation des protéines vers différentes membranes. Le génome d'Arabidopsis contient onze gènes homologues à PHO1. Neuf de ces homologues sont répartis en trois groupes. PHO1 et PHO1;H1 forment un de ces groupes. Nos travaux ont démontré que seuls PHO1;H1 et PHO1, sous contrôle du promoteur PHO1, peuvent complémenter le mutant pho1. PHO1 est exprimé principalement dans le cylindre vasculaire de la racine et faiblement dans la partie aérienne. Le degré d'expression de PHO1;H1 est similaire dans le cylindre vasculaire de la racine et des feuilles. Ceci suggère que PHO1;H1 est aussi impliqué dans la mobilisation du Pi dans les feuilles, en plus de son rôle dans le transfert du Pi dans le xylème des racines. Afin de mieux explorer le rôle de PHO1, la dynamique du phosphate a été observée dans trois lignées de plantes transgéniques: un sur-expresseur de PHO1, un sous-expresseur de PHO1 et le mutant pho1. La sur-expression de PHO1 dans le tissue vasculaire des feuilles a provoqué l'efflux du Pi vers l'espace apoplastic du xylème, ce qui confirme le rôle de PHO1 dans le chargement du Pi dans le xylème. La sur-expressìon de PHO1 dans la rosette est responsable d'un changement de la dynamique du Pi et de la diminution de la croissance, ce qui fut démontré par une expérience de greffe de la rosette du sur-expresseur de PHO1 sur les racines du sauvage. On a observé pour le sous-expresseur de PHO1 et le mutant pho1 une diminution du phosphore d'environ 2 fais au niveau des feuilles, et une diminution de 5-10 fois du Pi dans les vacuoles des feuilles, par rapport au sauvage. Ceci confirme le rôle proposé de PHO1 dans le transfert du Pi des racines aux feuilles. La carence de Pi chez pho1 implique une diminution de la taille de la rosette. Pour expliquer ce phénotype une autre expérience de greffe démontra que la cause de ce changement provenait des racines. Ceci renforce l'hypothèse de l'importance du rôle de PHO1 dans le xylème de la racine pour le chargement du Pi. Le mutant phot et le sous-expresseur de PHO1 accumulent 8-15 fois plus de Pi dans leurs racines comparé au sauvage. Cependant, contrairement au phot mutant, le sous-expresseur de PHO1 avait une croissance comparable au sauvage malgré le niveau bas du Pi dans les feuilles. Ceci suggère que la taille de la rosette lors d'une carence en Pi chez Arabidopsis serait la conséquence d'un changement de concentration de Pi dans les racines ou d'une influence de la protéine PHO1.

Enhancing and diminishing gene function in human embryonic stem cells

It is widely recognized that gain- and loss-of-function approaches are essential for understanding the functions of specific genes, and such approaches would be particularly valuable in studies involving human embryonic stem (hES) cells. We describe a simple and efficient approach using lipofection to transfect hES cells, which enabled us to generate hES cell lines expressing naturally fluorescent green or red proteins without affecting cell pluripotency. We used these cell lines to establish a means of diminishing gene function using small interfering (si)RNAs, which were effective at knocking down gene expression in hES cells. We then demonstrated that stable expression of siRNA could knock down the expression of endogenous genes. Application of these gain- and loss-of-function approaches should have widespread use, not only in revealing the developmental roles of specific human genes, but also for their utility in modulating differentiation.

Relationship between global structural parameters and Enzyme Commission hierarchy: Implications for function prediction

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Gene expression in placentation of farm animals: An overview of gene function during development

Eutherian mammals share a common ancestor that evolved into two main placental types, i.e., hemotrophic (e.g., human and mouse) and histiotrophic (e.g., farm animals), which differ in invasiveness. Pregnancies initiated with assisted reproductive techniques (ART) in farm animals are at increased risk of failure; these losses were associated with placental defects, perhaps due to altered gene expression. Developmentally regulated genes in the placenta seem highly phylogenetically conserved, whereas those expressed later in pregnancy are more species-specific. To elucidate differences between hemotrophic and epitheliochorial placentae, gene expression data were compiled from microarray studies of bovine placental tissues at various stages of pregnancy. Moreover, an in silico subtractive library was constructed based on homology of bovine genes to the database of zebrafish - a nonplacental vertebrate. In addition, the list of placental preferentially expressed genes for the human and mouse were collected using bioinformatics tools (Tissue-specific Gene Expression and Regulation [TiGER] - for humans, and tissue-specific genes database (TiSGeD) - for mice and humans). Humans, mice, and cattle shared 93 genes expressed in their placentae. Most of these were related to immune function (based on analysis of gene ontology). Cattle and women shared expression of 23 genes, mostly related to hormonal activity, whereas mice and women shared 16 genes (primarily sexual differentiation and glycoprotein biology). Because the number of genes expressed by the placentae of both cattle and mice were similar (based on cluster analysis), we concluded that both cattle and mice were suitable models to study the biology of the human placenta. (C) 2011 Elsevier B.V. All rights reserved.

Relationship between global structural parameters and Enzyme Commission hierarchy: Implications for function prediction

In protein databases there is a substantial number of proteins structurally determined but without function annotation. Understanding the relationship between function and structure can be useful to predict function on a large scale. We have analyzed the similarities in global physicochemical parameters for a set of enzymes which were classified according to the four Enzyme Commission (EC) hierarchical levels. Using relevance theory we introduced a distance between proteins in the space of physicochemical characteristics. This was done by minimizing a cost function of the metric tensor built to reflect the EC classification system. Using an unsupervised clustering method on a set of 1025 enzymes, we obtained no relevant clustering formation compatible with EC classification. The distance distributions between enzymes from the same EC group and from different EC groups were compared by histograms. Such analysis was also performed using sequence alignment similarity as a distance. Our results suggest that global structure parameters are not sufficient to segregate enzymes according to EC hierarchy. This indicates that features essential for function are rather local than global. Consequently, methods for predicting function based on global attributes should not obtain high accuracy in main EC classes prediction without relying on similarities between enzymes from training and validation datasets. Furthermore, these results are consistent with a substantial number of studies suggesting that function evolves fundamentally by recruitment, i.e., a same protein motif or fold can be used to perform different enzymatic functions and a few specific amino acids (AAs) are actually responsible for enzyme activity. These essential amino acids should belong to active sites and an effective method for predicting function should be able to recognize them. (C) 2012 Elsevier Ltd. All rights reserved.

Disruption of vitellogenin gene function in adult honeybees by intra-abdominal injection of double-stranded RNA

Abstract Background The ability to manipulate the genetic networks underlying the physiological and behavioural repertoires of the adult honeybee worker (Apis mellifera) is likely to deepen our understanding of issues such as learning and memory generation, ageing, and the regulatory anatomy of social systems in proximate as well as evolutionary terms. Here we assess two methods for probing gene function by RNA interference (RNAi) in adult honeybees. Results The vitellogenin gene was chosen as target because its expression is unlikely to have a phenotypic effect until the adult stage in bees. This allowed us to introduce dsRNA in preblastoderm eggs without affecting gene function during development. Of workers reared from eggs injected with dsRNA derived from a 504 bp stretch of the vitellogenin coding sequence, 15% had strongly reduced levels of vitellogenin mRNA. When dsRNA was introduced by intra-abdominal injection in newly emerged bees, almost all individuals (96 %) showed the mutant phenotype. An RNA-fragment with an apparent size similar to the template dsRNA was still present in this group after 15 days. Conclusion Injection of dsRNA in eggs at the preblastoderm stage seems to allow disruption of gene function in all developmental stages. To dissect gene function in the adult stage, the intra-abdominal injection technique seems superior to egg injection as it gives a much higher penetrance, it is much simpler, and it makes it possible to address genes that are also expressed in the embryonic, larval or pupal stages.

Learning with Kernels on Graphs: DAG-based kernels, data streams and RNA function prediction.

In many application domains data can be naturally represented as graphs. When the application of analytical solutions for a given problem is unfeasible, machine learning techniques could be a viable way to solve the problem. Classical machine learning techniques are defined for data represented in a vectorial form. Recently some of them have been extended to deal directly with structured data. Among those techniques, kernel methods have shown promising results both from the computational complexity and the predictive performance point of view. Kernel methods allow to avoid an explicit mapping in a vectorial form relying on kernel functions, which informally are functions calculating a similarity measure between two entities. However, the definition of good kernels for graphs is a challenging problem because of the difficulty to find a good tradeoff between computational complexity and expressiveness. Another problem we face is learning on data streams, where a potentially unbounded sequence of data is generated by some sources. There are three main contributions in this thesis. The first contribution is the definition of a new family of kernels for graphs based on Directed Acyclic Graphs (DAGs). We analyzed two kernels from this family, achieving state-of-the-art results from both the computational and the classification point of view on real-world datasets. The second contribution consists in making the application of learning algorithms for streams of graphs feasible. Moreover,we defined a principled way for the memory management. The third contribution is the application of machine learning techniques for structured data to non-coding RNA function prediction. In this setting, the secondary structure is thought to carry relevant information. However, existing methods considering the secondary structure have prohibitively high computational complexity. We propose to apply kernel methods on this domain, obtaining state-of-the-art results.

A ribozyme-mediated, gene “knockdown” strategy for the identification of gene function in zebrafish

The zebrafish system offers many unique opportunities for the study of molecular biology. To date, only random mutagenesis, and not directed gene knockouts, have been demonstrated in this system. To more fully develop the potential of the zebrafish system, an approach to effectively inhibit the expression of any targeted gene in the developing zebrafish embryo has been developed. This approach uses a transient, cytoplasmic, T7 expression system, injected into the fertilized zebrafish egg to rapidly produce high levels of a ribozyme directed against the mRNA encoded by the targeted gene to inhibit its expression. In a demonstration of this strategy, expression of the recessive dominant zebrafish no tail gene was effectively inhibited by using this strategy to yield a phenotype identical to that resulting from a known defective mutation in this same gene. This, ribozyme-mediated, message deletion strategy may have use in determining the function of genetic coding sequences of unknown function.

Gene discovery and gene function assignment in filamentous fungi

Filamentous fungi are a large group of diverse and economically important microorganisms. Large-scale gene disruption strategies developed in budding yeast are not applicable to these organisms because of their larger genomes and lower rate of targeted integration (TI) during transformation. We developed transposon-arrayed gene knockouts (TAGKO) to discover genes and simultaneously create gene disruption cassettes for subsequent transformation and mutant analysis. Transposons carrying a bacterial and fungal drug resistance marker are used to mutagenize individual cosmids or entire libraries in vitro. Cosmids are annotated by DNA sequence analysis at the transposon insertion sites, and cosmid inserts are liberated to direct insertional mutagenesis events in the genome. Based on saturation analysis of a cosmid insert and insertions in a fungal cosmid library, we show that TAGKO can be used to rapidly identify and mutate genes. We further show that insertions can create alterations in gene expression, and we have used this approach to investigate an amino acid oxidation pathway in two important fungal phytopathogens.

Functional genomics: Probing plant gene function and expression with transposons

Transposable elements provide a convenient and flexible means to disrupt plant genes, so allowing their function to be assessed. By engineering transposons to carry reporter genes and regulatory signals, the expression of target genes can be monitored and to some extent manipulated. Two strategies for using transposons to assess gene function are outlined here: First, the PCR can be used to identify plants that carry insertions into specific genes from among pools of heavily mutagenized individuals (site-selected transposon mutagenesis). This method requires that high copy transposons be used and that a relatively large number of reactions be performed to identify insertions into genes of interest. Second, a large library of plants, each carrying a unique insertion, can be generated. Each insertion site then can be amplified and sequenced systematically. These two methods have been demonstrated in maize, Arabidopsis, and other plant species, and the relative merits of each are discussed in the context of plant genome research.