Mutations in the RNA-binding protein TDP-43 cause amyotrophic lateral sclerosis (ALS). Inclusions enriched in TDP-43 in the cytoplasm of spinal cord neurons in both familial and sporadic ALS are hallmarks of the disease. Two recent yeast genome-wide loss-of-function toxicity suppressor screens revealed the strongest suppressor of TDP-43-mediated toxicity is the ablation of the gene encoding the metallophosphoesterase (MPE) Dbr1, the only enzyme known to hydrolyze the 2',5'-phosphodiester bonds formed within introns during their excision from pre-mRNA by the spliceosome. Decreasing Dbr1 activity results in the accumulation of RNA lariats that are proposed to sequester pathogenic TDP-43, preventing it from interfering with normal RNA metabolism. Supporting this hypothesis, knockdown of debranching activity in yeast, a human neuronal cell line, and in primary rat neurons protects them from TDP-43-mediated toxicity. The high degree of sequence identity in the catalytic domains of Dbr1 proteins across all eukaryotic species supports the observation that Dbr1's influence on TDP-43 activity is similar from yeast to man. We recently determined the first crystal structures of an RNA lariat debranching enzyme alone and in complex with a synthetic RNA containing a bona fide branchpoint identical to those found in intron RNA lariats. Dbr1 from the eukaryotic organism Entamoeba histolytica (Eh) crystallized most readily, revealing several unexpected features of Dbr1 enzymes relative to other MPE family members. All Dbr1 enzymes are mononuclear, possessing an invariant active site cysteine residue in the position of the aspartic acid observed in all previously characterized MPE superfamily members, all of which are active as dinuclear enzymes. In addition, all Dbr1 proteins contain a highly conserved insertion loop not found in other MPEs we term the lariat recognition loop (LRL). Functional data coming from in vivo complementation assays using multiple Dbr1 variants expressed in trans in dbr1 yeast support the proposed mononuclear enzymatic mechanism, as well as the roles assigned to the various unique structural elements observed in the crystal structures. The structures also reveal the molecular basis for how Dbr1 distinguishes 22,52-phosphodiester linkages from the far more abundant 32,52-phosphodiester linkages. With these results in-hand, we are now in possession of the tools needed to test the hypothesis that inhibition of Dbr1 represents a novel therapeutic avenue to treat TDP-43-mediated ALS. These tools include: 1) large quantities of purified Dbr1 proteins from multiple species, including human; 2) a tested, robust in vitro RNA debranching assay amenable to high throughput screening (HTS) of small molecule libraries for inhibitors; 3) the ability to synthesize, in parallel with the HTS inhibitor search, branched RNA analogs through the introduction of sugar modifications and linkers in order to mimic the conformation of the branched RNA we observe bound to the enzyme crystallographically, which differs from the conformations of these species in solution; 4) an in vivo complementation assay to test the effectiveness of potential inhibitors coming from both of the pipelines mentioned above; 5) the ability to rapidly observe the atomic details of inhibitorDbr1 complexes to enable the rational design of compounds with greater affinity and specificity; 6) the ability to perform initial toxiciy screens of candidate Dbr1 inhibitor compounds using cultured human neurons; and 7) the ability to pursue the structure of human Dbr1. The completion of the work outlined in this proposal on the Dbr1 enzyme is required before testing inhibitors in cell-based and murine models of TDP-43 mediated toxicity in ALS can begin.
U.S. veterans, particularly those deployed in the Persian Gulf War of 1991, show increased risk for developing ALS, a progressive neurodegenerative disease without cure or effective treatment. The one drug approved for ALS is Rilutek. If there is a benefit for ALS patients taking the drug, it may (in selected patients) extend survival by ~3 months. The primary reason that no effective therapies have been developed is that there is a paucity of 'druggable' target proteins known to affect the progression of the disease. This lack of protein targets in ALS is also a problem in other neurodegenerative disorders in which proteins are observed to aggregate. Recent data suggest that in the RNA lariat debranching enzyme Dbr1, there may finally be a druggable enzyme that when inhibited, could provide tangible benefit to veteran and non-veteran ALS patients alike. The experiments outlined in this proposal are designed to identify the first generation of small molecule Dbr1 inhibitors.
