Tsunehiro Mizushima

Abstract Dysregulation of DNA double-strand break (DSB) repair leads to adaptive immunodeficiency, whereas the remaining lymphocytes are aberrantly activated and provoke inflammations. However, no model mice were available to consistently manifest inflammation under defective DSB repair. We generated mutant mice carrying a missense mutation p.W447C in the gene encoding DNA ligase IV (LIG4), critical for DSB repair.Lig4^W447C/W447Cmice showed growth retardation and severe intestinal inflammations under adaptive immunodeficiency. The inflammations were featured by marked infiltration of T helper type 1 (Th1) cells and macrophages and was dependent on lymphocytes. WhenIfngwas deleted, Th2 and Th17 instead of Th1 cells drove the inflammations.Lig4^W447C/W447Cmice showed expansion of oligoclonal T cells with T cell receptor α repertoire skewed towards more proximal 3’ V and 5’ J gene segments. Thus, our novel hypomorphicLig4mutant mice show that defective DSB repair leads to Th1-dependent intestinal inflammations under severe adaptive immunodeficiency.

The cytosolic peptide:N-glycanase (PNGase) is involved in the quality control of N-glycoproteins via the endoplasmic reticulum-associated degradation (ERAD) pathway. Mutations in the gene encoding cytosolic PNGase (NGLY1 in humans) cause NGLY1 deficiency. Recent findings indicate that the F-box protein FBS2 of the SCFFBS2 ubiquitin ligase complex can be a promising drug target for NGLY1 deficiency. Here, we determined the crystal structure of bovine FBS2 complexed with the adaptor protein SKP1 and a sugar ligand, Man3GlcNAc2, which corresponds to the core pentasaccharide of N-glycan. Our crystallographic data together with NMR data revealed the structural basis of disparate sugar-binding specificities in homologous FBS proteins and identified a potential druggable pocket for in silico docking studies. Our results provide a potential basis for the development of selective inhibitors against FBS2 in NGLY1 deficiency.

Deficiencies in mitochondrial protein import are associated with a number of diseases. However, although nonimported mitochondrial proteins are at great risk of aggregation, it remains largely unclear how their accumulation causes cell dysfunction. Here, we show that nonimported citrate synthase is targeted for proteasomal degradation by the ubiquitin ligase SCFUcc1. Unexpectedly, our structural and genetic analyses revealed that nonimported citrate synthase appears to form an enzymatically active conformation in the cytosol. Its excess accumulation caused ectopic citrate synthesis, which, in turn, led to an imbalance in carbon flux of sugar, a reduction of the pool of amino acids and nucleotides, and a growth defect. Under these conditions, translation repression is induced and acts as a protective mechanism that mitigates the growth defect. We propose that the consequence of mitochondrial import failure is not limited to proteotoxic insults, but that the accumulation of a nonimported metabolic enzyme elicits ectopic metabolic stress.

Summary Pathogenic bacteria deliver virulence factors called effectors into host cells in order to facilitate infection. The Shigella effector proteins IpaH1.4 and IpaH2.5 are members of the “novel E3 ligase” (NEL)-type bacterial E3 ligase family. These proteins ubiquitinate the linear ubiquitin assembly complex (LUBAC) to inhibit nuclear factor (NF)-κB activation and, concomitantly, the inflammatory response. However, the molecular mechanisms underlying the interaction and recognition between IpaH1.4 and IpaH2.5 and LUBAC is unclear. Here we present the crystal structures of the substrate-recognition domains of IpaH1.4 and IpaH2.5 at resolutions of 1.4 and 3.4 Å, respectively. The LUBAC-binding site on IpaH1.4 was predicted based on structural comparisons with the structures of other NEL-type E3s. Structural and biochemical data were collected and analyzed to determine the specific residues of IpaH1.4 that are involved in interactions with LUBAC and influence NF-κB signaling. The new structural insight presented here demonstrates how bacterial pathogens target innate immune signaling pathways.

Impaired proteasome activity due to genetic variants of certain subunits might lead to proteasome-associated autoinflammatory syndromes (PRAAS). Here we report a de novo heterozygous missense variant of the PSMB9 proteasome subunit gene in two unrelated Japanese infants resulting in amino acid substitution of the glycine (G) by aspartic acid (D) at position 156 of the encoded protein β1i. In addition to PRAAS-like manifestations, these individuals suffer from pulmonary hypertension and immunodeficiency, which are distinct from typical PRAAS symptoms. The missense variant results in impaired immunoproteasome maturation and activity, yet ubiquitin accumulation is hardly detectable in the patients. A mouse model of the heterozygous human genetic variant (Psmb9G156D/+) recapitulates the proteasome defects and the immunodeficiency phenotype of patients. Structurally, PSMB9 G156D interferes with the β-ring-βring interaction of the wild type protein that is necessary for 20S proteasome formation. We propose the term, proteasome-associated autoinflammatory syndrome with immunodeficiency (PRAAS-ID), to indicate a separate category of autoinflammatory diseases, similar to, but distinct from PRAAS, that describes the patients in this study.

The Keap1-Nrf2 system is central for mammalian cytoprotection against various stresses and a drug target for disease prevention and treatment. One model for the molecular mechanisms leading to Nrf2 activation is the Hinge-Latch model, where the DLGex-binding motif of Nrf2 dissociates from Keap1 as a latch, while the ETGE motif remains attached to Keap1 as a hinge. To overcome the technical difficulties in examining the binding status of the two motifs during protein-protein interaction (PPI) simultaneously, we utilized NMR spectroscopy titration experiments. Our results revealed that latch dissociation is triggered by low-molecular-weight Keap1-Nrf2 PPI inhibitors and occurs during p62-mediated Nrf2 activation, but not by electrophilic Nrf2 inducers. This study demonstrates that Keap1 utilizes a unique Hinge-Latch mechanism for Nrf2 activation upon challenge by non-electrophilic PPI-inhibiting stimuli, and provides critical insight for the pharmacological development of next-generation Nrf2 activators targeting the Keap1-Nrf2 PPI.

Autophagy contributes to the selective degradation of liquid droplets, including the P-Granule, Ape1-complex and p62/SQSTM1-body, although the molecular mechanisms and physiological relevance of selective degradation remain unclear. In this report, we describe the properties of endogenous p62-bodies, the effect of autophagosome biogenesis on these bodies, and the in vivo significance of their turnover. p62-bodies are low-liquidity gels containing ubiquitin and core autophagy-related proteins. Multiple autophagosomes form on the p62-gels, and the interaction of autophagosome-localizing Atg8-proteins with p62 directs autophagosome formation toward the p62-gel. Keap1 also reversibly translocates to the p62-gels in a p62-binding dependent fashion to activate the transcription factor Nrf2. Mice deficient for Atg8-interaction-dependent selective autophagy show that impaired turnover of p62-gels leads to Nrf2 hyperactivation in vivo. These results indicate that p62-gels are not simple substrates for autophagy but serve as platforms for both autophagosome formation and anti-oxidative stress.

The transmembrane intracellular lectin ER-Golgi intermediate compartment protein 53 (ERGIC-53) and the soluble EF-hand multiple coagulation factor deficiency protein 2 (MCFD2) form a complex that functions as a cargo receptor, trafficking various glycoproteins between the endoplasmic reticulum (ER) and the Golgi apparatus. It has been demonstrated that the carbohydrate-recognition domain (CRD) of ERGIC-53 (ERGIC-53CRD) interacts with N-linked glycans on cargo glycoproteins, whereas MCFD2 recognizes polypeptide segments of cargo glycoproteins. Crystal structures of ERGIC-53CRD complexed with MCFD2 and mannosyl oligosaccharides have revealed protein-protein and protein-sugar binding modes. In contrast, the polypeptide-recognition mechanism of MCFD2 remains largely unknown. Here, a 1.60 Å resolution crystal structure of the ERGIC-53CRD-MCFD2 complex is reported, along with three other crystal forms. Comparison of these structures with those previously reported reveal that MCFD2, but not ERGIC-53-CRD, exhibits significant conformational plasticity that may be relevant to its accommodation of various polypeptide ligands.

Moyamoya disease (MMD) is a cerebrovascular disease characterized by progressive occlusion of the internal carotid arteries. Genetic studies originally identified RNF213 as an MMD susceptibility gene that encodes a large 591 kDa protein with a functional RING domain and dual AAA+ ATPase domains. As the functions of RNF213 and its relationship to MMD onset are unknown, we set out to characterize the ubiquitin ligase activity of RNF213, and the effects of MMD patient mutations on these activities and on other cellular processes. In vitro ubiquitination assays, using the RNF213 RING domain, identified Ubc13/Uev1A as a key ubiquitin conjugating enzyme that together generate K63-linked polyubiquitin chains. However, nearly all MMD patient mutations in the RING domain greatly reduced this activity. When full-length proteins were overexpressed in HEK293T cells, patient mutations that abolished the ubiquitin ligase activities conversely enhanced nuclear factor κB (NFκB) activation and induced apoptosis accompanied with Caspase-3 activation. These induced activities were dependent on the RNF213 AAA+ domain. Our results suggest that the NFκB- and apoptosis-inducing functions of RNF213 may be negatively regulated by its ubiquitin ligase activity and that disruption of this regulation could contribute towards MMD onset.

Citrate synthase (CS) catalyzes the conversion of oxaloacetate and acetyl coenzyme A into citrate and coenzyme A in the mitochondrial tricarboxylic acid (TCA) cycle. In plants, mitochondrial metabolism, including the TCA cycle, occurs in interaction with photosynthetic metabolism. The controlled regulation of several enzymes in the TCA cycle, such as CS, is important in plants. Here, the first crystal structure of a plant mitochondrial CS, CSY4 from Arabidopsis thaliana (AtCSY4), has been determined. Structural comparison of AtCSY4 with mitochondrial CSs revealed a high level of similarity. Inhibition analysis showed a similar manner of inhibition as in mitochondrial CSs. The effect of oxidation on one of a pair of cysteine residues in AtCSY4 was speculated upon based on the folded structure.

Eukaryotic proteasome assembly is a highly organized process mediated by several proteasome-specific chaperones, which interact with proteasome assembly intermediates. In yeast, Ump1 and Pba1-4 have been identified as assembly chaperones that are dedicated to the formation of the proteasome 20S catalytic core complex. The crystal structures of Pba chaperones have been reported previously, but no detailed information has been provided for the structure of Ump1. Thus, to better understand the mechanisms underlying Ump1-mediated proteasome assembly, we characterized the conformation of Ump1 in solution using NMR. Backbone chemical shift data indicated that Ump1 is an intrinsically unstructured protein and largely devoid of secondary structural elements.

Eukaryotic proteasome assembly is assisted by multiple dedicated chaperones. In yeast, formation of the heteroheptameric ring composed of α1-α7 subunits is promoted by the heterodimeric chaperone Pba3-Pba4. Here we reveal that in the absence of this dimeric chaperone, α2 replaces α4 during α-ring assembly, thereby giving rise to a non-productive complex that lacks α4, β1, β5, β6, and β7 subunits and aggregates of α4. Furthermore, our structure-guided mutational data demonstrate that the Pba3-Pba4 heterodimer acts as molecular matchmaker reinforcing the interaction between α4 and α5, which is the crucial step in the α-ring formation.

Proteasome formation does not occur due to spontaneous self-organization but results from a highly ordered process assisted by several assembly chaperones. The assembly of the proteasome ATPase subunits is assisted by four client-specific chaperones, of which three have been structurally resolved. Here, we provide the structural basis for the working mechanisms of the last, hereto structurally uncharacterized assembly chaperone, Nas2. We revealed that Nas2 binds to the Rpt5 subunit in a bivalent mode: the N-terminal helical domain of Nas2 masks the Rpt1-interacting surface of Rpt5, whereas its C-terminal PDZ domain caps the C-terminal proteasome-activating motif. Thus, Nas2 operates as a proteasome activation blocker, offering a checkpoint during the formation of the 19S ATPase prior to its docking onto the proteolytic 20S core particle.

Assembly of the eukaryotic 20S proteasome is an ordered process involving several proteins operating as proteasome assembly factors including PAC1-PAC2 but archaeal 20S proteasome subunits can spontaneously assemble into an active cylindrical architecture. Recent bioinformatic analysis identified archaeal PAC1-PAC2 homologs PbaA and PbaB. However, it remains unclear whether such assembly factor-like proteins play an indispensable role in orchestration of proteasome subunits in archaea. We revealed that PbaB forms a homotetramer and exerts a dual function as an ATP-independent proteasome activator and a molecular chaperone through its tentacle-like C-terminal segments. Our findings provide insights into molecular evolution relationships between proteasome activators and assembly factors.

HOIL-1L and its binding partner HOIP are essential components of the E3-ligase complex that generates linear ubiquitin (Ub) chains, which are critical regulators of NF-κB activation. Using crystallographic and mutational approaches, we characterize the unexpected structural basis for the specific interaction between the Ub-like domain (UBL) of HOIL-1L and the Ub-associated domain (UBA) of HOIP. Our data indicate the functional significance of this non-canonical mode of UBA-UBL interaction in E3 complex formation and subsequent NF-κB activation. This study highlights the versatility and specificity of protein-protein interactions involving Ub/UBLs and their cognate proteins.

Removal of the fucose residue from the N-glycans of the Fc portion of immunoglobulin G (IgG) results in a dramatic enhancement of antibody-dependent cellular cytotoxicity (ADCC) through improved affinity for Fcγ receptor IIIa (FcγRIIIa). Here, we present the 2.2-Å structure of the complex formed between nonfucosylated IgG1-Fc and a soluble form of FcγRIIIa (sFcγRIIIa) with two N-glycosylation sites. The crystal structure shows that one of the two N-glycans of sFcγRIIIa mediates the interaction with nonfucosylated Fc, thereby stabilizing the complex. However, fucosylation of the Fc N-glycans inhibits this interaction, because of steric hindrance, and furthermore, negatively affects the dynamics of the receptor binding site. Our results offer a structural basis for improvement in ADCC of therapeutic antibodies by defucosylation.