Chaperone-mediated autophagy (HSPA8 substrate delivery and LAMP-2A translocation)

Chaperone-mediated autophagy (CMA) is a selective lysosomal degradation pathway for soluble cytosolic proteins that carry a KFERQ-like pentapeptide motif. The constitutive Hsp70 chaperone HSPA8/Hsc70 recognizes the motif in the cytosol and delivers the substrate to the cytosolic side of the lysosomal membrane, where it binds monomeric LAMP-2A, the CMA-specific splice isoform of LAMP2. Substrate binding drives LAMP-2A multimerization into a high-molecular-weight translocation complex, the unfolded substrate crosses the membrane, and it is degraded by lumenal hydrolases; HSPA8 then disassembles the LAMP-2A complex, returning LAMP-2A to its monomeric, substrate-binding state. The module covers substrate recognition and delivery, translocation through the LAMP-2A complex, and complex recycling. Regulation of CMA (for example lysosomal mTORC2/PHLPP1/AKT signalling or cathepsin A-mediated LAMP-2A turnover), the cochaperones that assist HSPA8 at the membrane, and the downstream consequences of degrading particular substrates (NLRP3, BBC3/PUMA, GPX4) are outside the boundary.

Built as the decomposition target for the precomposed GO:0061740 "protein targeting to lysosome involved in chaperone-mediated autophagy" (projects/INVOLVED_IN_PRECOMPOSITION.md): the delivery step is GO:0006622 protein targeting to lysosome as part of GO:0061684 chaperone-mediated autophagy. Human proteins are representative members (HSPA8 P11142, reviewed in genes/human/HSPA8; LAMP2 P13473, isoform LAMP-2A P13473-1 per UniProt, no gene review yet). The HSPA8 annoton function, protein-macromolecule adaptor activity, matches both the HSPA8 core function and the HSPA8 activity in production GO-CAM gomodel:63c0ac2b00001252. That model types the LAMP2 activity as signaling adaptor activity (GO:0035591), which describes a signal relay rather than substrate binding and translocation. The cited data show LAMP-2A binds the substrate and that its multimerization is required for translocation, but not whether LAMP-2A itself forms the conduit. Neither transmembrane protein transporter activity (the conduit claim) nor cargo receptor activity (defined for delivery to transport vesicles) is established, so the LAMP-2A function carries no GO id, only a preferred term. No PANTHER family ids are asserted.