BACKGROUND: Selection on proteins is typically measured with the assumption that each protein acts independently. However, selection more likely acts at higher levels of biological organization, requiring an integrative view of protein function. Here, we built a kinetic model for de novo pyrimidine biosynthesis in the yeast Saccharomyces cerevisiae to relate pathway function to selective pressures on individual protein-encoding genes. RESULTS: Gene families across yeast were constructed for each member of the pathway and the ratio of nonsynonymous to synonymous nucleotide substitution rates (dN/dS) was estimated for each enzyme from S. cerevisiae and closely related species. We found a positive relationship between the influence that each enzyme has on pathway function and its selective constraint. CONCLUSIONS: We expect this trend to be locally present for enzymes that have pathway control, but over longer evolutionary timescales we expect that mutation-selection balance may change the enzymes that have pathway control.

The most widely used classification system describing enzyme-catalysed reactions is the Enzyme Commission (EC) number. Understanding enzyme function is important for both fundamental scientific and pharmaceutical reasons. The EC classification is essentially unrelated to the reaction mechanism. In this work we address two important questions related to enzyme function diversity. First, to investigate the relationship between the reaction mechanisms as described in the MACiE (Mechanism, Annotation, and Classification in Enzymes) database and the main top-level class of the EC classification. Second, how well these enzymes biocatalysis are adapted in nature. In this thesis, we have retrieved 335 enzyme reactions from the MACiE database. We consider two ways of encoding the reaction mechanism in descriptors, and three approaches that encode only the overall chemical reaction. To proceed through my work, we first develop a basic model to cluster the enzymatic reactions. Global study of enzyme reaction mechanism may provide important insights for better understanding of the diversity of chemical reactions of enzymes. Clustering analysis in such research is very common practice. Clustering algorithms suffer from various issues, such as requiring determination of the input parameters and stopping criteria, and very often a need to specify the number of clusters in advance. Using several well known metrics, we tried to optimize the clustering outputs for each of the algorithms, with equivocal results that suggested the existence of between two and over a hundred clusters. This motivated us to design and implement our algorithm, PFClust (Parameter-Free Clustering), where no prior information is required to determine the number of cluster. The analysis highlights the structure of the enzyme overall and mechanistic reaction. This suggests that mechanistic similarity can influence approaches for function prediction and automatic annotation of newly discovered protein and gene sequences. We then develop and evaluate the method for enzyme function prediction using machine learning methods. Our results suggest that pairs of similar enzyme reactions tend to proceed by different mechanisms. The machine learning method needs only chemoinformatics descriptors as an input and is applicable for regression analysis. The last phase of this work is to test the evolution of chemical mechanisms mapped onto ancestral enzymes. This domain occurrence and abundance in modern proteins has showed that the / architecture is probably the oldest fold design. These observations have important implications for the origins of biochemistry and for exploring structure-function relationships. Over half of the known mechanisms are introduced before architectural diversification over the evolutionary time. The other halves of the mechanisms are invented gradually over the evolutionary timeline just after organismal diversification. Moreover, many common mechanisms includes fundamental building blocks of enzyme chemistry were found to be associated with the ancestral fold.

Synthesis of the CCA end of essential tRNAs is performed either by CCA-adding enzymes or as a collaboration between enzymes restricted to CC- and A-incorporation. While the occurrence of such tRNA nucleotidyltransferases with partial activities seemed to be restricted to Bacteria, the first example of such split CCA-adding activities was reported in Schizosaccharomyces pombe. Here, we demonstrate that the choanoflagellate Salpingoeca rosetta also carries CC- and A-adding enzymes. However, these enzymes have distinct evolutionary origins. Furthermore, the restricted activity of the eukaryotic CC-adding enzymes has evolved in a different way compared to their bacterial counterparts. Yet, the molecular basis is very similar, as highly conserved positions within a catalytically important flexible loop region are missing in the CC-adding enzymes. For both the CC-adding enzymes from S. rosetta as well as S. pombe, introduction of the loop elements from closely related enzymes with full activity was able to restore CCA-addition, corroborating the significance of this loop in the evolution of bacterial as well as eukaryotic tRNA nucleotidyltransferases. Our data demonstrate that partial CC- and A-adding activities in Bacteria and Eukaryotes are based on the same mechanistic principles but, surprisingly, originate from different evolutionary events.

Today, enzymes are extensively used for many industrial applications, this includes bulk and fine-chemical synthesis, pharmaceuticals and consumer products. Though Nature has perfected enzymes for many millions of years, they seldom reach industrial performance targets. Natural enzymes could benefit from protein redesign experiments to gain novel functions or optimize existing functions. Glutathione transferases (GSTs) are detoxification enzymes, they also display disparate functions. Two Mu-class GSTs, M1-1 and M2-2, are closely related but display dissimilar substrate selectivity profiles. Saturation mutagenesis of a previously recognized hypervariable amino acid in GST M2-2, generated twenty enzyme variants with altered substrate selectivity profiles, as well as modified thermostabilities and expressivities. This indicates an evolutionary significance; GST Mu-class enzymes could easily alter functions in a duplicate gene by a single-point mutation. To further identify residues responsible for substrate selectivity in the GST M2-2 active site, three residues were chosen for iterative saturation mutagenesis. Mutations in position10, identified as highly conserved, rendered enzyme variants with substrate selectivity profiles resembling that of specialist enzymes. Ile10 could be conserved to sustain the broad substrate acceptance displayed by GST Mu-class enzymes. Enzymes are constructed from primarily twenty amino acids, it is a reasonable assumption that expansion of the amino acid repertoire could result in functional properties that cannot be accomplished with the natural set of building blocks. A combination approach of site-directed mutagenesis and chemical modifications in GST M2-2 and GST M1-1 resulted in novel enzyme variants that displayed altered substrate selectivity patterns as well as improved enantioselectivities. The results presented in this thesis demonstrate the use of different protein redesign techniques to modulate various functions in Mu-class GSTs. These techniques could be useful in search of optimized enzyme variants for industrial targets. / biokemi och organisk kemi

L’habileté des enzymes à évoluer joue un rôle fondamental dans l'adaptation des organismes à leur environnement, leur permettant de s'adapter aux changements de température, aux nutriments disponibles ou encore à l'introduction de composés cytotoxiques. Au cours des dernières décennies, cette capacité a conduit à l'émergence rapide de mécanismes de résistance aux antibiotiques chez des bactéries pathogènes pour l’humain, notamment dans le cas de l'antibiotique synthétique triméthoprime. Dix ans après l'introduction de cet antibiotique, l'enzyme dihydrofolate réductase de type B (DfrB) a été identifiée comme conférant une résistance aux bactéries l'exprimant en catalysant par voie d’enzyme alternative la réaction inhibée par l’antibiotique. Des études structurales, cinétiques et mécanistiques de la DfrB en ont révélé la nature atypique, et suggèrent que cette enzyme est un modèle d’enzyme primitive. En particulier, son site actif unique est formé via l’interface de quatre protomères identiques. Puisque les DfrB ne sont pas apparentées sur le plan évolutif à des protéines connues et caractérisées, on ne connait pas comment elles ont évolué pour ultimement contribuer à la résistance au triméthoprime, et en particulier comment leur capacité catalytique a émergé au sein du petit domaine codé par leurs gènes. Ainsi, cette thèse vise à approfondir notre compréhension de l’évolution des enzymes en examinant spécifiquement l’évolution des DfrB et les propriétés qui ont guidé ce processus. Puisque les gènes des DfrB ont rarement été rapportés, je présente d’abord nos efforts déployés pour identifier et caractériser de manière génomique les DfrB dans les bases de données publiques. Ces efforts ont conduit à la découverte, pour la première fois, de DfrB en dehors du contexte clinique. Nous avons ensuite caractérisé, sur le plan biophysique et enzymatique, des homologues protéiques aux DfrB que nous avons identifiés dans des bases de données de protéines putatives. Nous avons démontré la capacité d’homologues identifiés dans des contextes environnementaux, non associés aux activités humaines, à catalyser la réduction du dihydrofolate de la même façon que les DfrB. Enfin, une large exploration d’homologues de séquence, suivie d'une caractérisation expérimentale et computationnelle, nous a permis d'identifier des homologues distants des DfrB, certains capables de procurer une résistance au triméthoprime, et d'autres dépourvus de cette capacité. Ces résultats nous ont permis de proposer un modèle expliquant l’émergence de l'activité catalytique au sein du domaine protéique des DfrB. En résumé, cette thèse présente une approche multidisciplinaire pour l’exploration et la caractérisation de l’espace de séquence d’une famille de protéines. Cette approche, qui comprend des analyses génomiques, enzymologiques, biophysiques et bio-informatiques, nous a permis d’identifier les caractéristiques structurales et de séquences nécessaires à la formation d’une enzyme DfrB fonctionnelle. Nous avons également proposé un modèle pour expliquer l’évolution de cette enzyme primitive. Dans l’ensemble, nos résultats suggèrent que la capacité catalytique des DfrB a évolué indépendamment de l’introduction de l’antibiotique triméthoprime, et donc que ce mécanisme de résistance existait dans l’environnement préalablement à son recrutement génomique dans un contexte clinique. Ces travaux contribuent à notre compréhension fondamentale des mécanismes sous-jacents à l’émergence de l’activité catalytique au sein d’un domaine protéique non catalytique, et informent les études des mécanismes développés par les bactéries pour proliférer en présence d’antibiotiques. / The ability of enzymes to evolve plays a fundamental role in the adaptation of organisms to their environment, allowing them to adjust to changes in temperature, available nutrients, or the introduction of cytotoxic compounds. In recent decades, this ability has led to the rapid emergence of antibiotic resistance mechanisms in human pathogenic bacteria, particularly in the case of the synthetic antibiotic trimethoprim. Ten years after the introduction of this antibiotic, the type B dihydrofolate reductase (DfrB) was identified as conferring resistance to bacteria expressing it by providing an alternative enzyme to catalyze the reaction inhibited by the antibiotic. Structural, kinetic, and mechanistic studies of DfrB have revealed its atypical nature and suggest that this enzyme is a model of a primitive enzyme. In particular, its unique active site is formed by the interface of four identical protomers. Since DfrB enzymes are not evolutionarily related to any known and characterized proteins, it is not known how they evolved to ultimately contribute to trimethoprim resistance, and in particular how their catalytic ability arose within the small domain encoded by their genes. Thus, this thesis aims to deepen our understanding of enzyme evolution by specifically examining the evolution of DfrB and the properties that guided this process. Since DfrB genes have rarely been reported, I first present our efforts to genomically identify and characterize DfrB in public databases. These efforts led to the first discovery of DfrB genes outside the clinical context. We then biophysically and enzymatically characterized protein homologues of the DfrB we identified in putative protein databases. We demonstrated the ability of homologues identified in environmental contexts unrelated to human activities to catalyze dihydrofolate reduction in the same manner as DfrB. Finally, a broad search for sequence homologues, followed by experimental and computational characterization, allowed us to identify distant DfrB homologues, some capable of conferring resistance to trimethoprim and others lacking this ability. These results have allowed us to propose a model that explains the emergence of catalytic activity within the DfrB domain. In summary, this thesis presents a multidisciplinary approach to explore and characterize the sequence space of a protein family. This approach, which includes genomic, enzymatic, biophysical and bioinformatic analyses, has enabled us to identify the structural and sequence features necessary for the formation of a functional DfrB enzyme. We have also proposed a model to explain the evolution of this primitive enzyme. Overall, our results suggest that the catalytic capacity of DfrB evolved independently of the introduction of the antibiotic trimethoprim, and thus that this resistance mechanism existed in the environment prior to its genomic recruitment in a clinical context. This work contributes to our fundamental understanding of the mechanisms underlying the emergence of catalytic activity within a non-catalytic protein domain, and informs studies of the mechanisms developed by bacteria to proliferate in the presence of antibiotics.