|Home | About | Journals | Submit | Contact Us | Français|
Autophagic adapters including p62/SQSTM1 recognize polyubiquitinated autophagic targets such as toxic protein aggregates. Recently reporting in Molecular Cell, Filimonenko et al. provide evidence that the phosphatidylinositol 3-phosphate (PI3P) binding protein. Alfy interacts with p62, Atg5, and PI3P to coordinate target recognition with site-specific activation of autophagic components.
Autophagy represents a collection of interrelated cytoplasmic quality and quantity control systems that maintain cellular viability, primarily through sequestration and degradation of a wide range of cytoplasmic components that need to be removed to ensure proper cellular function and survival. With broad roles in cellular homeostasis, autophagy affects many aspects of human health and pathology, including normal processes such as aging and disease states such as cancer, metabolic illnesses, myopathies, infection and inflammation, and neurodegeneration (Mizushima et al., 2008). Physically, sensu stricto autophagy (referred to as macroautophagy) entails the capture of cytoplasmic targets into double membrane organelles termed autophagosomes that mature into degradative organelles termed autolysosomes. When regulating biomass quantity or responding to nutritional needs (e.g. during starvation or hypoxia), autophagy sequesters large sections of cytoplasm, leading to their turnover and conversion into energy and nutrients. In its quality control role, autophagy removes surplus, damaged, and often toxic cytoplasmic components ranging from protein aggregates too large for removal by the proteasomal system to nonmembranous organelles (e.g. ribosomes) to membrane-bound organelles such as depolarized or leaky mitochondira, excess preroxisomes, overproliferated ER, and even microbial intruders in the cytoplasm (Deretic, 2010).
Regulation and execution of autophagy is relatively well understood, primarily owing to the discovery of Atg factors in yeast and their counterparts in mammalian cells. The Atg factors run the core autophagic machinery in all eukaryotic cells from yeast to man. The hallmark of the transformation of a membrane into an autophagosome that elongates and wraps around its target is the presence of one of the Atg factors, Atg8, also known in mammalian cells as LC3 (more precisely as several members of the LC3/GABARAP family). The association of LC3 with a membrane is the consequence of Atg8 lipidation at its C-terminus with phosphatidylethanolamine (PE) (Figure 1); LC3-PE eventually decorates the nascent autophagosome. How do cells know what targets in the cytoplasm to capture with LC3-PE positive autophagosomes and commit to autophagic degradation? The answer to this question has been somewhat lagging behind the unraveling of the core Atg machinery. One of the first proteins recognized as an adapter for delivering cargo earmarked by polyubiquitination to the autophagic organelles is p62/SQSTM1 (Figure 1) (Bjorkoy et al., 2005). p62 bridges the cargo and autophagic machinery by binding to LC3 via its LIR (LC3-interacting region) motif (Figure 1) and binding to polyubiquitinated tags on cargo earmarked for autophagic degradation via its UBA domain. The targets for p62-dependent autophagy range in nature and size, and include protein aggregates (Bjorkoy et al., 2005), mitochondria polyubiqutinated by Parkin on VDAC1 (Geisler et al., 2010), and a gamut of intracellular microbes (Deretic, 2010) destined for degradation and elimination or ribosomal proteins converted into neoantimicrobial peptides in the innate immune functions of autophagy (Ponpuak et al., 2010). Since the characterization of p62, additional adapters such as NBR1 have been identified (Kirkin et al., 2009). However, their target and functional specificity or overlaps remain to be fully delineated, as both p62 and NBR1 display similar principal features. Filimonenko et al. (2010), in a recent issue of Molecular Cell, now describes a precise function for another type of autophagic adapter – the phosphatidylinositol 3-phosphate (PI3P) binding protein Alfy (autophagy linked FYVE protein) – that does not posses, at least not overtly, motifs seen in p62 and NBR1. Alfy interacts with or affects multiple Atg factors (Figure 1A) and lipids, including p62, the Atg12-Atg5-Atg16 complex (which serves as an E3-like enzyme to position and enhance LC3 lipidation into LC3-PE), and PI3P (via a lipid-binding domain called FYVE). The work of Filimonenko et al. suggests that Alfy may be a master conductor that orchestrates the assembly of autophagic organelles surrounding p62-captured cargo (Figure 1B).
Alfy was first characterized as a FYVE domain-containing, enormous protein of 400kDa found in association with protein inclusions and autophagosomes. It is normally localized to the nucleus, along the nuclear membrane, and colocalizes with nucleoporins. Alfy can be found on nuclear promyelocytic leukemia bodies (PMLs), where misfolded proteins accumulate. Although it is not required for starvation-induced autophagic degradation of long-lived proteins, Alfy shifts to cytoplasmic ubiquitin-positive aggregates upon stress, including starvation. The resting nuclear localization of Alfy may come at first blush as a surprise. However, Alfy’s nuclear localization and translocation into the cytoplasm is reminiscent of the similar p62 shuttling between the nucleus and the cytosol. Indeed, Alfy’s translocation to the cytosol depends on p62 (Clausen et al., 2010).
In their study, Filimonenko and colleagues (2010) overcame the unwieldy size of the Alfy protein (which makes its molecular manipulations and analyses difficult) and show, in an experimental tour de force, Alfy’s physiological and molecular function. Alfy protects against the neurotoxicity of aggregate-prone proteins such as the polyglutamine (polyQ) mutant Huntingtin, which is associated with Huntington’s disease, and α-synuclein, which is associated with Parkinson’s disease. The authors find that the Drosophila Alfy ortholog Blue Cheese is protective in a Drosophila eye model of polyQ disease, where it suppressed polyQ transgene effects of reduced eye size, loss of pigmentation, ommatidial disorganization, and necrosis. Both ends of the Alfy protein interacts with the polyQ target, and Alfy was found in immunoprecipitates with p62 (and NBR1) along with the mutant polyQ aggregates. The p62-interaction with the polyQ target was Alfy-independent, indicating that p62-dependent recognition of cargo earmarked for autophagy is probably the very initial event of the process of target capture for autophagy. Alfy colocalized with Atg5 and co-precipitated (via its WD40 domain) with a 56 kDa complex containing Atg5 covalently conjugated to Atg12 and non-covalently bound to Atg16L. Since the Atg5-Atg12-Atg16L complex governs the site and extent of LC3 lipidation, it is likely that Alfy first recognizes the p62-captured cargo and then stimulates LC3-PE production to bring about autophagic membrane recruitment or formation around the offending protein aggregate. As p62 (and NBR1) has an LC3-interacting motif (LIR), the gestalt of autophagy can be completed by Alfy’s capacity to stimulate LC3 lipidation to form LC3-PE, whereas p62 presents LC3 as a substrate for lipidation, finds and binds to LC3-PE that is already stimulated by Alfy-Atg5-Atg12-Atg16 action, or both.
Finally, one may ask what the Alfy PI3P-binding domain FYVE does for Alfy and for the entire autophagic process? PI3P and its synthesizing enzyme, the class III PI3 kinase hVPS34, are essential for autophagy, although the exact function of PI3P in autophagy is still elusive. There are a number of proteins that interact with PI3P that have been implicated in autophagy. These include Atg18 and its mammalian equivalents WIPI-1 and WIPI-2, which may play a role in membrane traffic to and from growing autophagosomes. PI3P also interacts with DFCP-1 (with two FYVE domains), an association that has no known role but is considered to be a marker for autophagosome formation sites at or in the vicinity of the ER. Lastly, PI3P interacts with FYCO, a FYVE protein that links autophagosomes to microtubule-associated motors (Pankiv et al.). It is possible to imagine that Alfy may either recognize one or more PI3P membranes at several of these stages in autophagy, perhaps even playing a role in membrane recruitment and growth around a p62-captured protein aggregate or other targets. It is also tempting to consider the possibility that Alfy may help transform the nature of membranes initially containing only PI3P by converting them into LC3-PE-positive membrane, thus gradually conferring upon them an autophagic identity and function. Regardless of the specifics, it is clear that the formation of autophagosomes that are productive in the removal of toxic aggregates is not a trivial process but represents the product of careful and layered orchestration by Alfy and p62 in the execution of aggregate recognition, capture and sequestration into autophagosomes.