TARDBP/TDP-43 regulates autophagy in both MTORC1-dependent and MTORC1-independent manners

ABSTRACT

In a recent paper we addressed the mechanism by which defective autophagy contributes to TARDBP/TDP-43-mediated neurodegenerative disorders. We demonstrated that TARDBP regulates MTORC1-TFEB signaling by targeting RPTOR/raptor, a key component and an adaptor protein of MTORC1. Loss of TARDBP decreased the mRNA stability of RPTOR and this regulation in turn enhanced autophagosomal and lysosomal biogenesis in an MTORC1-dependent manner. Meanwhile, loss of TARDBP could also impair autophagosome-lysosome fusion in an MTORC1-independent manner. Importantly, we found that modulation of MTOR activity by treatment with rapamycin and phosphatidic acid had strong effects on the neurodegenerative phenotypes of TBPH (Drosophila TARDBP)-depleted flies. Taken together, our data reveal that multiple dysfunctions in the autophagic process contribute to TARDBP-linked neurodegeneration and may help to identify potential therapeutic targets in the future.

TARDBP is a major pathogenic component of amyotrophic lateral sclerosis (ALS), frontotemporal dementia and other neurodegenerative diseases. TARDBP is a RNA-binding protein that has multiple functions including transcription, pre-mRNA splicing, mRNA transport and stability. Inappropriate cleavage and aggregation of TARDBP can be found in ALS and frontotemporal dementia patients' brains. Although the pathogenesis of TARDBP-mediated neurodegeneration is still not fully understood, increasing evidence shows that TARDBP loss of function is a major mediator in disease progress. The autophagy-lysosome pathway (ALP) is a protein quality control system, which is critical for cellular homeostasis. Neurodegenerative diseases are tightly associated with ALP, since dysfunctions in ALP have been observed in many cases of these diseases and mutations in ALP genes can cause neurodegeneration such as ALS. Here we summarize our recent study demonstrating that the overall function of ALP is affected in TARDBP-depleted models and how defective ALP leads to TARDBP-mediated neurodegeneration.

In our study, we identified the MTORC1 component RPTOR as a novel molecular target of TARDBP. We found that the RNA recognition motifs in TARDBP can directly interact with RPTOR mRNA, and that TARDBP depletion reduces mRNA stability of RPTOR. Hence, the mRNA and protein levels of RPTOR are both decreased in TARDBP-depleted cells. Our results further suggest that TARDBP depletion reduces MTOR lysosomal localization and MTORC1 activity, and induces downstream TFEB nuclear translocation through RPTOR, but not other components associated with MTORC1, such as Ragulator or RRAG GTPases.

Our results suggest that TARDBP depletion not only results in TFEB nuclear translocation, but also in TFEB lysosomal enrichment. We reasoned that in TARDBP-deficient cells, a decreased RPTOR level would lead to inactivation of MTOR, which in turn inhibits the association between TFEB and YWHA/14-3-3 in the cytosol, and enhances the association between TFEB and RRAG GTPases on the lysosome surface. Meanwhile, an inactivation of MTOR by TARDBP depletion decreases TFEB phosphorylation and increases its nuclear translocation, leading to an activation of ALP gene expressions. Relevant to this, we observed increased autophagosomal and lysosomal gene expressions in TARDBP-depleted cells, which in turn increases autophagosomal and lysosomal biogenesis.

Using electronic and confocal microscopy, we observed an abnormal accumulation of autophagosomes and lysosomes in TARDBP-deficient cells. Moreover, biochemical analysis also showed defective autophagic substrate turnover in those cells. These phenomena are similar to those observed in other ALS and neurodegenerative disease models, showing an accumulation of immature autophagic vesicles, which indicates that dysfunction of ALP (autophagy flux) may be associated with various types of diseases. We further revealed a MTORC1-independent mechanism by which TARDBP depletion overwhelms ALP function despite its also enhancing autophagosomal and lysosomal biogenesis through MTORC1-TFEB signaling. Our data suggest that DCTN1/dynactin 1, a motor protein that is involved in autophagosome-lysosome fusion, is a mediator of defective ALP driven by TARDBP depletion. TARDBP depletion impairs autophagosome-lysosome fusion in a DCTN1-dependent manner, indicating that an impairment of autophagosome-lysosome fusion is a cause of defective autophagic turnover and ALP function in TARDBP loss-of-function models.

Since the neuron is not capable of cell division and is vulnerable to cellular stress, and given that the biosynthesis of an autophagosome is faster and more efficient than autophagosome-lysosome fusion, abnormal accumulation of immature autophagic vesicles may trigger neurotoxicity in disease progression. Relevant to this, the overall ALP is like a highway with traffic. Under normal conditions, the traffic is "healthy" since the incoming vehicles (newly synthesized autophagic vesicles) can go through the entire highway rapidly (successful autophagosome-lysosome fusion). In the case of TARDBP loss of function, the incoming vehicles are increased but at the same time the highway is jammed (impaired autophagosome-lysosome fusion). Thus, the increased incoming traffic acts as a “stressor” and makes things even worse. Ultimately it makes all the autophagic vehicles "stuck" on the highway and the autophagic highway itself is overwhelmed. This model can explain our observations in Tbph knockout flies that an enhancement of autophagic vesicle biogenesis by the MTORC1 inhibitor rapamycin aggravates, whereas an inhibition of autophagic vesicle biogenesis by the MTORC1 agonist phosphatidic acid ameliorates, the neurotoxicity in Tbph knockout flies.

Taken together, our data highlight the important roles of MTORC1-TFEB signaling and MTORC1-independent blockage of autophagosome-lysosome fusion in TARDBP-linked neurodegeneration. Although the current view is that an induction of the earlier step of autophagic flux (MTORC1 inhibition) can exhibit therapeutic benefit to neurodegenerative disorders, our work suggests that treatments favoring the later step of autophagic flux (autophagosome-lysosome fusion) could be more efficient and helpful to survive neurons experiencing very stressful situations.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.