Understanding the Structural Basis for Human and Viral Nuclear Localisation Signal Domains with Importins and RNA

Eukaryotic cells are characterised by the presence of organelle boundaries. The nucleus, which contains genetic material, is one of the main organelles within the cell. Due to the cell membrane's impermeability to molecules larger than 40 kDa, cells have evolved mechanisms to transport macromolecules through it. Nuclear import relies significantly on translocation by nuclear import proteins. Proteins requiring entry into the nucleus possess specific domains called nuclear localisation signals (NLSs). This specific sequence enables proteins to bind to nuclear import proteins and gain access to the nucleus through nuclear pore complexes. The classical nuclear import pathway involves the importin α (IMPα) adaptor protein and the importin β (IMPβ1) import receptor protein, which form a nuclear import complex with the cargo, allowing it to enter the nucleus. Although classical NLSs commonly have multiple basic residues, the mechanism of interaction and the affinity of these sequences to classical import pathway proteins can vary. The primary objective of this study was to characterise the interaction between NLSs from human and viral proteins with classical nuclear import pathway proteins. Additionally, another part of the study investigated the interaction between the NLS domain of SOX transcription factors (TFs) and RNA.
The main goal of this thesis was to characterise the binding mechanism between NLS domains of human and viral proteins and classical nuclear import proteins. The first study investigated the interaction of the NLS domain in activator protein 1 (AP1) TFs with classical nuclear import proteins to address existing knowledge gaps in this area. By employing biochemical and biophysical assays, this study investigated the interaction between the basic leucine zipper (bZIP) domains of four AP1 TF members, including cJUN, cFOS, cMAF, and ATF2, and IMPα adaptor proteins. Using X-ray crystallography, this study determined the molecular interface between the NLSs of these AP1 members and IMPα2, which provides a better understanding of the various mechanisms these proteins utilise to interact with nuclear import proteins. These novel structures revealed a previously unidentified NLS at the N-terminus of the cJUN bZIP domain basic region, which was confirmed following mutational studies. Interestingly, we observed a unique binding mode between the cJUN NLS and IMPα2, which was not observed in structures of other AP1 members with IMPα2.
The second study investigated the NLS of the open reading frame 4b (ORF4b) viral accessory protein in Middle East respiratory syndrome coronavirus (MERS-CoV) and MERS-like viruses. The role of the MERS-CoV ORF4b NLS domain in suppressing the innate immune system through competitive binding to IMPα3 is well-established. However, this NLS domain in MERS-like viruses has not been comprehensively studied. Therefore, our aim was to investigate the interaction between the ORF4b NLS domains of four MERS-like viruses, including HKU4, HKU5, HKU25, and HKU31, and IMPα adaptor proteins, both biochemically and structurally. The qualitative and quantitative analyses revealed two basic sites in the NLS domains of MERS-like viruses, with varying importance. Furthermore, the structure of the HKU4 ORF4b NLS domain and IMPα2 revealed a monopartite binding fashion, similar to what was previously observed with the HKU5 ORF4b NLS. Interestingly, no interactions were observed between HKU31 ORF4b and any of the tested classical nuclear import pathway proteins. This data has provided new insights into the nuclear import of clinically significant viral accessory proteins and how they interact with nuclear import molecules in the cell.
It has recently been discovered that some TFs can interact not only with DNA but also with RNA. One such example is the SRY-related high mobility group (HMG) box (SOX) TFs, which have been shown to interact with RNA through their HMG-box domains. However, the exact interaction site has yet to be clearly identified. In the third study, we aimed to determine the region in the HMG-box of SOX proteins (one representative of each group) that mediates interaction with RNA, as well as the specific residues involved in this interaction, using various biochemical and biophysical techniques. Our experiments highlighted the importance of the C-terminal region of the SOX HMG-box in binding to RNA. Additionally, the significance of basic residues in this C-terminal region was demonstrated. Furthermore, we determined the structure of the HMG-box domain of the SOX30 protein, the only SOX protein unable to bind RNA and that lacks basic residues in its C-terminal region. Interestingly, despite being unable to bind RNA, SOX30 exhibited a structure similar to that of other SOX HMG-box proteins. This provides further evidence for the importance of the HMG-box domain C-terminal region in the RNA binding ability of SOX proteins.
Overall, this thesis presents comprehensive and novel information about the interaction of NLS domains in human and viral proteins with classical nuclear import pathway proteins, in addition to providing valuable insights into how NLS domains can also be involved in the RNA binding ability of SOX proteins. As the focus on targeting nuclear import mechanisms grows, these findings serve as a solid foundation for drug discovery efforts and the development of novel therapeutics.