Systemic deregulation of autophagy upon loss of ALS- and FTD-linked C9orf72

ABSTRACT

A genetic mutation in the C9orf72 gene causes the most common forms of neurodegenerative diseases amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). The C9orf72 protein, predicted to be a DENN-family protein, is reduced in ALS and FTD, but its functions remain poorly understood. Using a 3110043O21Rik/C9orf72 knockout mouse model, as well as cellular analysis, we have found that loss of C9orf72 causes alterations in the signaling states of central autophagy regulators. In particular, C9orf72 depletion leads to reduced activity of MTOR, a negative regulator of macroautophagy/autophagy, and concomitantly increased TFEB levels and nuclear translocation. Consistent with these alterations, cells exhibit enlarged lysosomal compartments and enhanced autophagic flux. Loss of the C9orf72 interaction partner SMCR8 results in similar phenotypes. Our findings suggest that C9orf72 functions as a potent negative regulator of autophagy, with a central role in coupling the cellular metabolic state with autophagy regulation. We thus propose C9orf72 as a fundamental component of autophagy signaling with implications in basic cell physiology and pathophysiology, including neurodegeneration.

A hexanucleotide repeat expansion in the promoter or intronic region of isoforms of the C9orf72 gene is responsible for the most common forms of the neurodegenerative diseases ALS and FTD, as well as rare cases of Alzheimer, Parkinson, Huntington, and other diseases. While the repeat expansion reduces the level of C9orf72 protein in patients, and insufficiency in protein function has been proposed to contribute to pathogenesis, the functions of the C9orf72 protein are poorly understood. C9orf72 is predicted to be a member of the DENN domain protein superfamily, whose members have been involved in RAB-mediated membrane trafficking pathways. Using animal, cell, and proteomic tools to study the function of the C9orf72 protein we report the surprising finding that loss of C9orf72 results in enhanced autophagic activity, under both basal and starvation conditions. Thus, we propose C9orf72 as a negative regulator of autophagy (Fig. 1).

Figure 1.

Impact of C9orf72 loss on autophagy regulation. (Left) Under normal growth conditions, MTORC1 inhibits both autophagy induction and nuclear translocation of TFEB, a master transcription factor for autophagy and lysosomal genes. (Right) In response to the loss of C9orf72, MTORC1 signaling is decreased, cellular TFEB levels and nuclear localization are increased, and lysosomal capacity and autophagic flux are enhanced.

We generated 3110043O21Rik/C9orf72 knockout mice and observed a late-onset age-dependent lethality phenotype in the homozygote, and to a lesser degree, the heterozygous animals. The cause of the animal death remains unclear, although immune dysregulation could be involved. To investigate the molecular functions of C9orf72 we used human and mouse knockdown/knockout cells, and determined that loss of the C9orf72 protein systemically perturbs autophagy regulation. We revealed an impairment to MTOR signaling because MTOR activation after amino acid stimulation is diminished in the absence of C9orf72. In addition, we observed an enhancement of TFEB (transcription factor EB) signaling: Loss of C9orf72 increases TFEB protein levels and promotes its nuclear translocation.

Consistent with inhibited MTOR and upregulated TFEB signaling, using western blot analysis we observed an increase in autophagic flux upon loss of C9orf72. This effect occurred under nutrient-rich conditions, and was more pronounced under nutrient starvation. To gain further insights we used quantitative microscopy to examine autophagic flux based on the numbers of LC3-positive autophagic vesicles and the colocalization between LC3 vesicles and RAB7, a late endosome and lysosome-associated GTPase that marks mature autolysosomes. Our analysis confirmed the increase in the autophagic flux capacity and enhanced autolysosome formation in C9orf72 knockout cells. Consistently, the autophagy marker protein SQSTM1/p62 was significantly decreased in the brain of C9orf72 knockout mice when the animals were fed with a low-protein diet. While initial reports proposed an opposite role of C9orf72 in impacting autophagic activities, our study demonstrates that C9orf72 is a negative regulator of autophagy. Recent reports support our findings that loss of C9orf72 increases autophagic flux and decreases MTOR signaling. Since autophagy is a dynamic process, the long-term effects of reduced C9orf72 functions on autophagy pathways in pathophysiological systems could be diverse and depend on specific cellular conditions.

We and others found that the most abundant interactor of C9orf72 protein is another DENN protein, SMCR8, and the 2 proteins likely form a cognate complex, since the depletion of each protein leads to significant decrease of the other. The most structurally homologous proteins to SMCR8 and C9orf72 in the human proteome are FLCN (folliculin) and FNIP (folliculin interacting protein), respectively, which are also DENN proteins. Similar to C9orf72 and SMCR8, FNIP and FLCN form a protein complex and regulate autophagy and MTOR signaling. We speculate that the C9orf72-SMCR8 complex, like the FNIP-FLCN complex, functions in regulating metabolic pathways.

The exact molecular function of C9orf72 and how it acts to regulate autophagy, MTOR, and TFEB signaling, needs to be further elucidated. One of the predicted functions of DENN proteins is the role of guanine nucleotide exchange factors for RAB GTPases. Interactions of C9orf72 with RAB1A, RAB8B, RAB39B, as well as RAB5, RAB7, and RAB11, have been reported; however, it still must be determined which RAB(s) is the major functional partner of C9orf72. Also, the exact site where C9orf72 acts in the regulation of metabolic signaling remains unclear. Autophagy, MTOR, and TFEB signaling are interconnected by organelle and protein networks of feedback loops that help regulate the metabolic demands of the cell. Given the major consequences on these pathways induced by loss of C9orf72 or SMCR8, further studies of the C9orf72-SMCR8 complex could lead to understanding of new metabolic regulatory processes still undiscovered in the cell.

The studies of C9orf72 and SMCR8 will also contribute to the understanding of human diseases such as C9orf72-linked ALS and FTD. Neurodegenerative diseases, including ALS, are increasingly associated with autophagy genes, such as UBQLN2 (ubiquilin 2), SQSTM1, OPTN (optineurin), and TBK1. Although the prevailing theory considers the impairment of autophagy as a cause for neurodegeneration, there have been surprising findings that activation of autophagy is insufficient to ameliorate neurodegeneration. Overall, our findings indicate that loss of C9orf72 derails metabolic regulation of autophagy, further highlighting the complexity linking autophagy and the relevant forms of neurodegeneration. A better understanding of C9orf72 will be useful for developing strategies to restore cellular homeostasis in diseased cells.

Funding

This work was supported by National Institutes of Health (NS074324 and NS089616), the Robert Packard Center for ALS Research at Johns Hopkins, and US Department of Defense (W81XWH-15-1-0069).