Structural and functional investigation of tRNA modifying enzymes: from simple methylation to complex wobble modification

Posttranscriptional tRNA modifications are essential for the proper functioning of tRNA in the frame of translation from mRNA to protein. These modifications can (i) contribute to the proper folding and stability of the tRNA, (ii) affect the aminoacylation properties, due to the recognition with the cognate aminoacyltransferase, and (iii) influence the correct codon - anticodon recognition at the decoding center of the ribosome. Even though most modifications present in tRNA were discovered 20-30 years ago, identification of tRNA modification genes has been lagging behind. However, advances in technology have led to the discovery of nearly 50 genes involved in tRNA modification since 2002. Not surprisingly considering the central role of tRNA and its modifications in translation, mutations in some of these genes have been linked to human diseases. To gain insight in the mechanisms of action of the enzymes corresponding to these newly discovered genes, detailed structural and biochemical/physical studies are required. In the framework of this PhD thesis we were focusing on three different tRNA modification enzymes and enzyme complexes, that are responsible for three different tRNA modifications, ranging from a simple methylation of guanosine (TrmN/Trm14) until complex multi-carbon modifications of wobble uridines (MnmEG complex and Elongator complex). TrmN/Trm14 (naming dependents on the kingdom of life) catalyzes the methylation of the exocyclic amino group of guanosine at position 6 of the tRNA, leading to an m2G6 modification. The protein consists of a THUMP domain, which is involved in RNA binding, linked to the catalytic Rossmann-fold methyltransferase (RFM) domain. In our study we solved crystal structures of this protein from two domains of life (i.e. Trm14 from the Archeon P. furiosus and TrmN from the bacterium T. thermophilus) in complex with different ligands. These crystal structures provide the first 3-dimensional views of proteins consisting of a THUMP domain linked to a RFM domain. We further focused on how these proteins bind to their tRNA substrates and especially on the way the THUMP domain - a proposed RNA-binding module - contributes to tRNA binding. We showed that the isolated THUMP domain of TrmN/Trm14 does not bind tRNA and the intact TrmN/Trm14 protein is needed for tRNA binding. This is in agreement with a docking model that we created for the tRNA-TrmN interaction. Here, the main binding groove for tRNA binding is formed by the interface of the RFM and THUMP domain, allowing the correct positioning of G6 onto the active site of the RFM domain. Modeling of the guanosine binding into the active site supports a flip out mechanism for G6, as seen for the rRNA methyltransferase RsmC. Future analysis will hopefully help to find out, how the THUMP domain is involved in the exact positioning of the tRNA substrate on the RFM domain to determine the specificity for modification of position 6 compared to other 155 methyltransferases consisting of the THUMP domain and RFM domain arrangement. In the second part we focused on the MnmEG complex, involved in the formation of the cmnm5 (5-carboxymethylaminomethyl) modification of wobble uridine (U34). While MnmG is the main responsible for tRNA binding in the modification reaction, MnmE is a GDP/GTP binding G protein activated by nucleotide dependent dimerization (GAD) that is probably involved in the regulation of the modification reaction. MnmE and MnmG were shown to form an a2b2 complex and both proteins supply co-factors (MTHF on MnmE; FAD on MnmG) to obtain cmnm5U34. Despite the fact that crystal structures of the isolated MnmE and MnmG proteins were bein