Hitomi Tsuiji

Summary Current gene silencing tools based on RNA interference (RNAi) or, more recently, clustered regularly interspaced short palindromic repeats (CRISPR)⃩Cas13 systems, have critical drawbacks, such as off-target effects (RNAi) or collateral mRNA cleavage (CRISPR⃩Cas13). Thus, a more specific method of gene knockdown is needed. Here, we developed “CRISPRδ”, an approach for translational silencing, harnessing catalytically inactive Cas13 proteins (dCas13). Owing to its tight association with mRNA, dCas13 serves as a physical roadblock for scanning ribosomes during translation initiation and does not affect mRNA stability. Guide RNAs covering the start codon lead to the highest efficacy regardless of the translation initiation mechanism: cap-dependent or internal ribosome entry site (IRES)-dependent translation. Strikingly, genome-wide ribosome profiling revealed the extremely high gene knockdown specificity of CRISPRδ. Moreover, fusion of a translational repressor to dCas13 ensured further improvement of the knockdown efficacy. Our method provides a framework for translational repression-based gene silencing in eukaryotes.

The polarity of the biological membrane, or lipid order, regulates many cellular events. It is generally believed that the plasma membrane polarity is regulated according to cell type and function, sometimes even within a cell. Neurons have a variety of functionally specialized subregions, each of which bears distinct proteins and lipids, and the membrane polarity of the subregions may differ accordingly. However, no direct experimental evidence of it has been presented to date. In the present study, we used a cell-impermeable solvatochromic membrane probe NR12A to investigate the local polarity of the plasma membrane of neurons. Both in hippocampal and cerebellar granule neurons, growth cones have higher membrane polarity than the cell body. In addition, the overall variation in the polarity value of each pixel was greater in the growth cone than in cell bodies, suggesting that the lateral diffusion and/or dynamics of the growth cone membrane are greater than other parts of the neuron. These tendencies were much less notably observed in the lamellipodia of a non-neuronal cell. Our results suggest that the membrane polarity of neuronal growth cones is unique and this characteristic may be important for its structure and function.

A GGGGCC repeat expansion in the C9ORF72 gene is the most common genetic cause of amyotrophic lateral sclerosis (ALS), a devastating motor neuron disease. In the neurons of ALS patients, dipeptide repeat proteins (DPRs) are produced from repeat-containing RNAs by an unconventional form of translation, and some of these proteins, especially those containing poly(glycine-arginine) and poly(proline-arginine), are toxic to neurons. Gemini of coiled bodies (GEMs) are nuclear structures that harbor survival of motor neuron (SMN) protein, and SMN is essential for the assembly of U-rich small nuclear ribonucleoproteins (snRNPs) that are central for splicing. We previously reported that GEMs are lost and that snRNP biogenesis is misregulated in the motor neurons of ALS patients. Here we show that DPRs interfere with GEM formation and proper SMN localization in HeLa cells and iPSC-derived motor neurons from an ALS patient with the C9ORF72 mutation. The accumulation of poly(glycine-arginine) markedly reduced the number of GEMs and caused the formation of aberrant cytoplasmic RNA granules that sequestered SMN. These findings indicate the functional impairment of SMN in motor neurons expressing DPRs and may provide a mechanism to explain the vulnerability of motor neurons of C9ORF72-ALS patients.

The RNA-binding protein Ataxin-2 binds to and stabilizes a number of mRNA sequences, including that of the transactive response DNA-binding protein of 43 kDa (TDP-43). Ataxin-2 is additionally involved in several processes requiring translation, such as germline formation, long-term habituation, and circadian rhythm formation. However, it has yet to be unambiguously demonstrated that Ataxin-2 is actually involved in activating the translation of its target mRNAs. Here we provide direct evidence from a polysome profile analysis showing that Ataxin-2 enhances translation of target mRNAs. Our recently established method for transcriptional pulse-chase analysis under conditions of suppressing deadenylation revealed that Ataxin-2 promotes post-transcriptional polyadenylation of the target mRNAs. Furthermore, Ataxin-2 binds to a poly(A)-binding protein PABPC1 and a noncanonical poly(A) polymerase PAPD4 via its intrinsically disordered region (amino acids 906-1095) to recruit PAPD4 to the targets. Post-transcriptional polyadenylation by Ataxin-2 explains not only how it activates translation but also how it stabilizes target mRNAs, including TDP-43 mRNA. Ataxin-2 is known to be a potent modifier of TDP-43 proteinopathies and to play a causative role in the neurodegenerative disease spinocerebellar ataxia type 2, so these findings suggest that Ataxin-2-induced cytoplasmic polyadenylation and activation of translation might impact neurodegeneration (i.e. TDP-43 proteinopathies), and this process could be a therapeutic target for Ataxin-2-related neurodegenerative disorders.

Reelin is a secreted protein that plays versatile roles in neuronal development and function. The strength of Reelin signaling is regulated by proteolytic processing, but its importance in vivo is not yet fully understood. Here, we generated Reelin knock-in (PA-DV KI) mice in which the key cleavage site of Reelin was abolished by mutation. As expected, the cleavage of Reelin was severely abrogated in the cerebral cortex and hippocampus of PA-DV KI mice. The amount of Dab1, whose degradation is induced by Reelin signaling, decreased in these tissues, indicating that the signaling strength of Reelin was augmented. The brains of PA-DV KI mice were largely structurally normal, but unexpectedly, the hippocampal layer was disturbed. This phenotype was ameliorated in hemizygote PA-DV KI mice, indicating that excess Reelin signaling is detrimental to hippocampal layer formation. The neuronal dendrites of PA-DV KI mice had more branches and were elongated compared to wild-type mice. These results present the first direct evidence of the physiological importance of Reelin cleavage.

Reelin plays important roles in regulating neuronal development, modulating synaptic function, and counteracting amyloid β toxicity. A specific proteolytic cleavage (N-t cleavage) of Reelin abolishes its biological activity. We recently identified ADAMTS-3 (a disintegrin and metalloproteinase with thrombospondin motifs 3) as the major N-t cleavage enzyme in the embryonic and early postnatal brain. The contribution of other proteases, particularly in the postnatal brain, has not been demonstrated in vivo. ADAMTS-2, -3 and -14 share similar domain structures and substrate specificity, raising the possibility that ADAMTS-2 and -14 may cleave Reelin. We found that recombinant ADAMTS-2 protein expressed in cultured cell lines cleaves Reelin at the N-t site as efficiently as ADAMTS-3 while recombinant ADAMTS-14 hardly cleaves Reelin. The disintegrin domain is necessary for the Reelin-cleaving activity of ADAMTS-2 and -3. ADAMTS-2 is expressed in the adult brain at approximately the same level as ADAMTS-3. We generated ADAMTS-2 knockout (KO) mice and found that ADAMTS-2 significantly contributes to the N-t cleavage and inactivation of Reelin in the postnatal cerebral cortex and hippocampus, but much less in the cerebellum. Therefore, it was suggested that ADAMTS-2 can be a therapeutic target for adult brain disorders such as schizophrenia and Alzheimer's disease.

Intracellular mislocalization of TAR DNA-binding protein 43 (TDP-43), a nuclear DNA/RNA-binding protein involved in RNA metabolism, is a pathological hallmark of amyotrophic lateral sclerosis (ALS). Although the aggregation-prone, TDP-43 C-terminal domain is widely considered as a key component of TDP-43 pathology in ALS, recent studies including ours suggest that TDP-43 N-terminal fragments (TDP-∆C) may also contribute to the motor dysfunction in ALS. However, the specific pathological functions of TDP-43 N-terminal fragments in mice have not been elucidated. Here, we established TDP-∆C knock-in mice missing a part of exon 6 of murine Tardbp gene, which encodes the C-terminal region of TDP-43. Homozygous TDP-∆C mice showed embryonic lethality, indicating that the N-terminal domain of TDP-43 alone is not sufficient for normal development. In contrast, heterozygous TDP-∆C mice developed normally but exhibited age-dependent mild motor dysfunction with a loss of C-boutons, large cholinergic synaptic terminals on spinal α-motor neurons. TDP-∆C protein broadly perturbed gene expression in the spinal cords of aged heterozygous TDP-∆C mice, including downregulation of Notch1 mRNA. Moreover, the level of Notch1 mRNA was suppressed both by TDP-43 depletion and TDP-∆C expression in Neuro2a cells. Decreased Notch1 mRNA expression in aged TDP-∆C mice was associated with the age-dependent motor dysfunction and loss of Akt surviving signal. Our findings indicate that the N-terminal region of TDP-43 derived from TDP-∆C induces the age-dependent motor dysfunction associated with impaired Notch1-Akt axis in mice.

Reelin is a secreted protein that antagonizes the deposition and toxicity of amyloid β peptide (Aβ). Therefore, augmentation of Reelin activity may ameliorate Alzheimer's disease (AD). We have recently reported that a disintegrin and metalloproteinase with thrombospondin motifs 3 (ADAMTS-3) cleaves and inactivates Reelin in the mouse brain. In the present study, we investigated the effect of reducing ADAMTS-3 on deposition of Aβ by crossbreeding drug-inducible ADAMTS-3 conditional knock-out (cKO) mice with "next-generation" AD model mice. We found that reducing ADAMTS-3 inhibited deposition of Aβ significantly in AppNL-F mice, which produce human wild-type Aβ. On the other hand, reducing ADAMTS-3 had no effect in AppNL-G-F mice, which produce the Arctic mutant Aβ (E22G) that forms protofibrils more efficiently than does wild-type Aβ. Thus, the findings suggest that the administration of an inhibitor against ADAMTS-3 will prevent the progression of AD pathology caused by deposition of wild-type Aβ.

Reelin is a secreted protein essential for the development and function of the mammalian brain. The receptors for Reelin, apolipoprotein E receptor 2 and very low-density lipoprotein receptor, belong to the low-density lipoprotein receptor family, but it is not known whether Reelin is involved in the brain lipid metabolism. In the present study, we performed lipidomic analysis of the cerebral cortex of wild-type and Reelin-deficient (reeler) mice, and found that reeler mice exhibited several compositional changes in phospholipids. First, the ratio of phospholipids containing one saturated fatty acid (FA) and one docosahexaenoic acid (DHA) or arachidonic acid (ARA) decreased. Secondly, the ratio of phospholipids containing one monounsaturated FA (MUFA) and one DHA or ARA increased. Thirdly, the ratio of phospholipids containing 5,8,11-eicosatrienoic acid, or Mead acid (MA), increased. Finally, the expression of stearoyl-CoA desaturase-1 (SCD-1) increased. As the increase of MA is seen as an index of polyunsaturated FA (PUFA) deficiency, and the expression of SCD-1 is suppressed by PUFA, these results strongly suggest that the loss of Reelin leads to PUFA deficiency. Hence, MUFA and MA are synthesized in response to this deficiency, in part by inducing SCD-1 expression. This is the first report of changes of FA composition in the reeler mouse brain and provides a basis for further investigating the new role of Reelin in the development and function of the brain.

Amyotrophic lateral sclerosis (ALS) is a late-onset motor neuron disorder. Although its neuropathology is well understood, the cellular and molecular mechanisms are yet to be elucidated due to limitations in the currently available human genetic data. In this study, we generated induced pluripotent stem cells (iPSC) from two familial ALS (FALS) patients with a missense mutation in the fused-in sarcoma (FUS) gene carrying the heterozygous FUS H517D mutation, and isogenic iPSCs with the homozygous FUS H517D mutation by genome editing technology. These cell-derived motor neurons mimicked several neurodegenerative phenotypes including mis-localization of FUS into cytosolic and stress granules under stress conditions, and cellular vulnerability. Moreover, exon array analysis using motor neuron precursor cells (MPCs) combined with CLIP-seq datasets revealed aberrant gene expression and/or splicing pattern in FALS MPCs. These results suggest that iPSC-derived motor neurons are a useful tool for analyzing the pathogenesis of human motor neuron disorders.