In eukaryotic cells short-lived proteins are degraded in a specific process by the ubiquitin-proteasome system (UPS), whereas long-lived proteins and damaged organelles are degraded by macroautophagy (hereafter referred to as autophagy). A growing body of evidence now suggests that autophagy is important for clearance of protein aggregates that form in cells as a consequence of ageing, oxidative stress, alterations that elevate the amounts of certain aggregation-prone proteins or expression of aggregating mutant variants of specific proteins. Autophagy is generally considered to be a non-specific, bulk degradation process. However, a recent study suggests that p62/SQSTM1 may link the recognition of polyubiquitinated protein aggregates to the autophagy machinery.1 This protein is able to polymerize via its N-terminal PB1 domain and to recognize polyubiquitin via its C-terminal UBA domain. It can also recruit the autophagosomal protein LC3 and co-localizes with many types of polyubiquitinated protein aggregates.1 Here we discuss possible implications of these findings and raise some questions for further investigation.