Autophagy is a self-digesting mechanism in which cells use lysosomes to degrade damaged or denatured macromolecular substances/organelles under the external environmental influence. In mammalian cells, there are three primary types of autophagy: microautophagy, macroautophagy, and chaperone-mediated autophagy (CMA).

During macroautophagy, double-membrane vesicles of autophagosomes deliver damaged proteins/organelles to lysosomes for degradation. Usually, autophagy mentioned in the literature refers to macroautophagy. Microautophagy transports substances to lysosomes for degradation through vesicles, where the lysosomal membrane is invaginated.

Unlike the first two types of autophagy, chaperone-mediated autophagy does not use vesicles for delivery. CMA is highly selective and often uses the chaperone protein HSC70 to degrade target proteins with a unique five-peptide motif (KFERQ-like). The receptor protein LAMP2A on the lysosomal membrane recognizes the exposed KFERQ group of the binding protein.,and then the target protein is recruited into the lysosome for degradation.

In neurodegenerative diseases, a significant amount of proteins that cannot be degraded accumulate in neurons. Degrading proteins through lysosomal autophagy is the primary way for cells to remove abnormal proteins. CMA has been reported to be involved in the degradation of pathogenic proteins in neurodegenerative disease. such as α-syn and tau; however, the consequences of the loss of function of the CMA in neurodegenerative disease remain unknown.

In 2021, Bourdenx M et al. published their study in Cell entitled Chaperone-mediated autophagy prevents collapse of the neuronal metastable proteome, revealing the critical regulatory role of CMA on neuronal homeostasis. This article provides an overview and further commentary on the neural homeostasis regulation study.

In this article, the researchers constructed a LAMP2A (L2A) knockout model (L2A^-/-) and a neuronal-specific L2A/LAMP2A (CKL2A^-/-) deletion mouse model to analyze the role of CMA in maintaining neuronal protein homeostasis (Fig. 3A). They also established a CKATG7^-/- mouse model that mimics the loss of macroautophagy (conditional knockout of ATG7)(Fig. 3B) and compared this with the CKL2A^-/- mouse model to explore the specific functions of the CMA and macroautophagy on neuronal protein degeneration (The mice without a knockout in the CTR group were the control group) ^[2].

At the beginning of the study, the researchers found that CKL2A^-/- and L2A^-/- mice showed higher scores on behavioral tests and faster hindlimb clenching progression compared to controls (CTR) (Fig. 4A). The mice also exhibited sensorimotor dysfunction and decreased short-term memory in the Y-maze (Fig. 4B-C). Only L2A^-/- mice exhibited Parkinson's disease-like gait characteristics, such as reduction in stride length (Fig. 4D). In addition, CKL2A^-/- mice also exhibited common neurodegeneration phenotypes , such as decreased spatial working memory and a trend of an apparent reduction in nesting behavior(Fig. 4E-F)^[2]. These results showed that neuronal L2A-deficient mice displayed most of the behavioral output of L2A systemically deficient mice.

The researchers found an accumulation of lipofuscin and K63-linked ubiquitinated proteins inclusions (both targeted for lysosomal degradation) in the hippocampus of L2A^−/− mice at 6 months of age. And similar features were found in excitatory pyramidal neurons in the hippocampus of CKL2A^-/- mice (Fig. 5A-B). Meanwhile, immunoblot analysis of CKL2A^-/- mice cortex showed an accumulation of oxidized ubiquitinated proteins (Figure 5C-D)^[2]. All these results concluded that CMA deletion led to the disruption of neuronal proteostasis.

And then the researchers performed quantitative proteomic analysis by isolating the sarkosyl-insoluble fraction from the cortex of CTR and CKL2A^-/- mice. The results showed that CKL2A^-/- mouse aggregated of many insoluble proteins in the brain, and 76% of these proteins contained KFERQ-like motifs. Proteins with KFERQ-like motifs, such as α-syn, tau, UCHL1, and PARK7, increased conversion to insoluble proteins in CKL2A^-/- mouse (Fig. 6B). Bourdenx M et al. also analyzed the supersaturation scoreof the proteins and found a 20.37-fold increase in insoluble proteins in CKL2A^-/- mice^[2].

The above results indicated that the proteome that becomes insoluble upon CMA blockage is part of the supersaturated proteome. Moreover, neuronal CMA blockage results in disruption of the proteome.

In the following step, Bourdenx M et al. compared proteomics to elucidate the specific roles of CMA and macroautophagy on neuronal protein degeneration in CKATG7^-/- and CKL2A^-/- mouse. Gene set enrichment and enrichment map analyses of proteins in the insoluble fractions (Fig. 7A) revealed that the altered proteome was not identical in CKATG7^-/- mice and CKL2A^-/- mice. Macroautophagy caused protein changes associated with the cell cycle and ubiquitinated proteasomal catabolic processes (blue arrows), whereas CMA deficiency was associated with protein trafficking and metabolism (red arrows). Measurement of extracellular acidification rate (ECAR) showed a marked reduction in glycolysis in CKL2A^-/- neurons (Fig. 7B). Moreover, more glycolytic enzymes increased in the insoluble fraction of CKL2A^-/- mice such as pyruvate dehydrogenase (PDH) (Fig. 7C)^[2].

Although macroautophagy blockade has been reported to affect neuronal glycolysis, the experimental results showed that L2A and ATG7 knockdown had different effects on glycolytic properties. The above results suggested that the outcome of the proteome of CMA-deficient neurons to insoluble proteins may differ from macroautophagy’s^[2]. Based on above evidence, it is suggested that direct blockade of neuronal CMA leads to the accumulation of insoluble proteins and alterations in neuronal function, which may increase vulnerability to neurodegenerative diseases and accelerate the progression of the diseases.

The researchers established a KFERQ-Dendra-hTau^P301L mice model and found that the CMA activity of neurons was reduced, and the number of CMA spots was significantly decreased(Fig. 8A). They also established a P301S tau transgenic mice model, in which CA77.1 (a derivative of AR7) was used to activate CMA in vitro without affecting macroautophagy (Fig. 8B). In vivo, CA77.1 administration normalized the PS19 mice (Fig. 8C). It significantly reduced the neurons containing pathogenic tau conformations in the hippocampus, amygdala, and piriform cortex (Fig. 8D).

The study demonstrated that CMA plays a vital role in regulating neuronal homeostasis. They used mice models with systemic and neuron-specific CMA blockade to prove that the absence of neuronal CMA results in protein toxicity and neuronal dysfunction. CMA deficiency also converts otherwise aggregation-prone KFERQ-like motif proteins into insoluble proteins. A neuron-specific CMA and ATG7-null mouse model was utilized to identify that CMA and macroautophagy are distinct from the subproteome of neurodegeneration in regulating neuronal proteostasis.

The article comprehensively clarified the role of CMA in the occurrence and development of neurodegenerative diseases, which has important guiding significance for future clinical treatment and drug development.