A loss-of-function variant in MLKL increases Alzheimer's disease risk through modulating [gamma]-secretase activity and autophagy

Abstract

Alzheimer’s disease (AD) is an age-related complex neurodegenerative disorder, which causes the most common type of dementia in human. Genetic factors play important roles in AD development. It has been demonstrated that pathogenic mutations in APP, PSEN1 and PSEN2 cause early-onset Alzheimer’s disease (EOAD). More than 95% AD cases were found in people over 65 years old, which is termed to be as late-onset AD (LOAD). In recent years, a variety of LOAD risk genes have been reported through Genome-wide Association Studies (GWAS), in which APOE ε4 is viewed as the most well-recognized genetic risk factor for LOAD. People carrying two copies of APOE ε4 allele have over 10 times higher LOAD risk than those without APOE ε4, while other genetic risk factors have small effect size. It has been reported that only about 40% of LOAD cases are not APOE ε4 carriers. In order to identify novel genetic factors with relatively higher effect size other than APOE ε4 that contribute to LOAD development, we performed a whole-exome sequencing analysis in APOE ε4-negative AD patients and matched controls from Hong Kong Chinese population. We recruited 246 APOE ε4-negative LOAD patients and 172 matched control subjects with normal cognition and whole-exome sequencing was performed. After bioinformatic analysis, a rare loss-of-function variant (p.Q48X) was found in gene MLKL (mixed lineage kinase domain like pseudokinase), which exclusively occurred in LOAD patients. Moreover, the results were replicated in other sample sets. The mutant (MLKL:p.Q48X) introduces a premature stop codon at the 48th codon, which results in nonsense-mediated decay of MLKL mRNA. In addition, my results indicated that Aβ42/Aβ40 ratio was significantly elevated in vitro at the absence of MLKL, while the phosphorylation of Tau was not altered by the mutation. To elucidate the mechanisms how Aβ generation is affected by the mutation, I firstly investigated the expression levels of APP and the secretases. The data showed that the expression levels of APP and the secretases have no changes in response to MLKL knock down (KD). Then I examined the activity of β- and γ-secretase, which mediate the production of Aβ, after MLKL KD or inhibition. The results demonstrated that β-secretase activity was not affected when MLKL was compromised. However, γ-secretase activity was found to be markedly up-regulated after MLKL was knock down or inhibited based on indirect and direct evidence. Moreover, my results indicated that the enhanced γ-secretase activity by the identified variant is likely to be attributed to the elevated formation of γ-secretase complex on membrane after MLKL was compromised. Collectively, my study identified a rare MLKL loss-of-function variant (MLKL:p.Q48X) in a Hong Kong Chinese APOE ε4-negative LOAD cohort, and this variant was demonstrated to confer susceptibility to AD through modulating γ-secretase activity. Autophagy is an evolutionarily conserved cellular event to maintain protein homeostasis. It has been reported that autophagy plays critical roles in multiple physiological and pathological events. Recently, accumulating evidence has been provided to closely link autophagy to AD pathogenesis. Based on the observation that MLKL localizes in the cytoplasm where autophagy is taking place, I explored the effects of the MLKL mutation on autophagy. My results showed that the accumulation of autophagosome was markedly compromised by MLKL KD or inhibition. In addition, the deficient autophagosome accumulation was demonstrated to result from inhibited autophagic flux instead of stimulated autophagosome degradation. As the mTORC1-ULK1 signaling was not altered in response to MLKL KD, I investigated the downstream event, VPS34 complex-mediated autophagy nucleation. The results indicated that the formation of VPS34 complex was compromised at the absence of MLKL. Subsequently, I revealed that Rab5, one of the upstream regulators of VPS34 complex, was significantly down-regulated after MLKL was knock down. In line with previous studies, our results showed that autophagosome accumulation was compromised via the VPS34 complex in response to Rab5 KD. Taken together, my study indicated that autophagosome formation was inhibited by the MLKL mutation and the process was mediated by the small GTPase Rab5. In conclusion, the present study identified a rare loss-of-function variant in MLKL (MLKL:p.Q48X), and the variant increases the risk to develop AD through modulating γ-secretase and autophagy.