Identification and Functional Characterization of the Docking Sites of FAT10 and NUB1L at the 26S Proteasome

Abstract

FAT10 is a 18 kDa protein with two ubiquitin-like domains encoded in the major histocompatibility complex class I region close to the HLA-F gene and hence, the assignment of its name as, HLA-F locus adjacent transcript 10. It is the only known ubiquitin-like modifier that targets its substrates for proteasomal degradation independently of ubiquitin. It bears a diglycine motif at the C-terminus that is required for the covalent conjugation with the target proteins destined for destruction. The degradation of FAT10 and its conjugates is further accelerated by its non-covalent interaction with an UBL-UBA domain protein NEDD8 ultimate buster-1 long (NUB1L). This process requires the attachment of NUB1L to the proteasome through its UBL domain although its interaction with FAT10 is not essential. Several studies provide hints about the significance of FAT10 in immune responses, including the synergistic induction of FAT10 in the presence of the proinflammatory cytokines (TNF-α and IFN-γ), its upregulation in several types of cancers, the hypersensitivity of the FAT10 knock-out mouse to lipopolysaccharide, high expression of FAT10 in the lymphoid organs such as thymus, lymph nodes and spleen, and its up-regulation upon maturation of dendritic cells.The primary goal of this thesis was to attain a deep knowledge on the mechanism of the degradation of FAT10 and its conjugates by the 26S proteasome. The first step was the identification of the subunit(s) of the proteasome with which FAT10 and NUB1L can interact. Yeast two hybrid and GST-pull down assays revealed a direct interaction of the hRpn10 subunit with FAT10. Unexpectedly, the N-terminal von Willebrand A (VWA) domain of hRpn10 bound to both FAT10 and NUB1L in contrast to ubiquitin, which binds ubiquitin interacting motifs (UIM1 and UIM2) in hRpn10. This finding raised the following questions: why and how NUB1L accelerates the degradation of FAT10, which are dealt with to some extent in this thesis. In addition, NUB1L can interact with Rpn1 subunit, apart from the known hRpn10 subunit. To understand the mechanism ofdegradation, a heterologous yeast system was employed because of the fact that yeast, but not mammalian cells, can survive, and show only minimal sensitivity to canavanine, in the absence of Rpn10. The VWA domain of hRpn10 can functionally reconstitute the rpn10Δ strain of yeast, as demonstrated by the degradation of FAT10. Moreover, it could be shown that this degradation is indeed dependent on the proteasome and not on the free cytosolic hRpn10. The importance of the VWA domain in the degradation of FAT10 was further highlighted by identifying a single amino acid residue, Asp11, which was critical for the degradation of FAT10. Interestingly, NUB1L is incapable of accelerating the degradation of FAT10 in the absence of hRpn10, which revealed an essential role of hRpn10 in the regulation of this pathway. This led to the identification of the VWAdomain as a novel interaction site for the ubiquitin-like proteins on the proteasome thatassists in the degradation of FAT10.Furthermore, by obtaining a three-dimensional structure of hRpn10 computationally, we tried to identify the docking site of FAT10 and NUB1L on the VWA domain of hRpn10. With the help of the three-dimensional structure of NUB1L generated by computational methods, a new putative UBA domain has been identified that was left unrecognized in earlier studies. The multiple sequence alignment of all the UBA domains of NUB1L and the UBA domains of a ubiquitin receptor, hHR23A, supported this finding. Furthermore, the VWA domain and the full-length hRpn10 proteins were purified for the future X-ray crystallography studies to understand more precisely the importance of the VWA domain in the context of protein degradation and prove our findings.Unlike ubiquitin, FAT10 is not a very stable protein and is degraded along with the covalently bound substrates. The biological relevance of FAT10, in the context of its yet unknown substrates, still remains enigmatic although its function has been several times related to apoptosis and cell cycle regulation. FAT10 mRNA is highly expressed in the lymphatic organs like thymus, spleen and lymph nodes but barely expressed in the brain. It is also known that FAT10 localizes to aggresomes under proteasome inhibition and this process is mediated by its interaction with HDAC6, a dynein motor complex associated protein. This finding encouraged us to determine the expression and significance of FAT10 in neurodegenerative diseases, which are characterized by the presence of aggregates in the brain. In this thesis, the up-regulation of FAT10 mRNA and protein is determined in the brain, after infection with LCMV and also in certain mouse models of neurodegenerative diseases, like Parkinson’s disease, Huntington’s disease and Alzheimer’s disease. Immunohistochemical stainings revealed the accumulation of FAT10 in aggregates in such diseases, which could suggest a protective role of FAT10 analogous to the function of ubiquitin in neurodegenerative diseases. Furthermore, to support this idea, monoclonal and polyclonal antibodies specific for mouse FAT10 were generated to study the FAT10 expression in the mouse models of different diseases by employing different proteomics approaches.In addition, this thesis deals with the FAT10-conjugation pathway, which is analogous to the ubiquitin conjugation pathway represented by a cascade of three enzymes: activating enzyme (E1), conjugating enzyme (E2) and ligation enzyme (E3). A novel E2 enzyme USE1 was identified, which accepts FAT10 as well as ubiquitin as substrates and is specific for the UBA6 E1 enzyme. UBA6-activated FAT10 can be transferred to USE1 in vitro and in vivo. Moreover, catalytically inactive USE1 was unable to interact with FAT10, which supports the characterization of USE1 as an E2 enzyme for FAT10. The siRNA mediated knock-down of USE1 down-regulated the conjugates of FAT10. Moreover, USE1 was identified as the first substrate of FAT10 as it can be auto- FAT10ylated in cis by an isopeptide bond. This provided a regulatory mechanism for the FAT10-conjugation pathway.In conclusion, this thesis provides detailed insights into the “FAT10-proteasome” and the “FAT10-conjugation” pathways. It could be shown that the expression of FAT10 is differentially regulated at several steps by hRpn10 and NUB1L. The VWA domain of hRpn10 is identified as a novel interaction site for ubiquitin-like modifiers and defined as a domain which can support the degradation of FAT10. This marks a major difference in the regulation of proteins controlled by the ubiquitin-proteasome and the “FAT10-proteasome” pathway. The conjugation studies provide further support to identify the covalently-modified substrates of FAT10, and the physiological role of FAT10. Moreover, the unexpected finding of the up-regulation and accumulation of FAT10 in aggregates in the brain of mouse models of neurodegenerative diseases would be beneficial in understanding these diseases by identifying the role of FAT10 and providing a significant link between FAT10 and the immune system.