Fig. 1. Ubiquitin, isopeptide linkage and chain conformation variations. (A) Ribbon diagram of the main chain of the ubiquitin molecule (Protein Data Bank ID: 1UBQ) with the seven lysine side chains and Ile44 shown in stick representation. Ubiquitin is composed of a single antiparallel -sheet curved over an adjacent -helix. Solvent-exposed Ile44 demarcates the chief contact surface of the ubiquitin molecule (Hicke et al., 2005; Hurley et al., 2006), and the majority of the ubiquitin lysine residues are located in a circumferential band relative to this key residue. Lys63, however, is positioned opposite the extended C terminus and Gly76 necessary for isopeptide linkage to a target lysine side chain. (B) Schematic representation of monoubiquitylation, Lys48-linked or Lys63-linked polyubiquitylation of target lysine residues. Note that the differential positioning of the lysine within the acceptor ubiquitin dictates different polyubiquitin chain conformations and geometries. We use the term oligoubiquitylation to indicate the presence of either multiple monoubiquitins, a polyubiquitin chain(s), or a combination of these modifications. (C) Examples of different ubiquitin configurations in plasma membrane proteins internalized in mammalian cells. (a) multiple monoubiquitylation (dimeric EGF receptors, for example), (b) polyubiquitylation with Lys63 linkage (for example, dimeric NGF receptor TrkA), (c) monoubiquitylation of subunits of homooligomeric complexes (for example, ROMK1), (d) polyubiquitylation of heterooligomeric complexes (for example, ENaC), (e) oligoubiquitylation in trans (for example, -arrestin 1/2 in complex with ₂ adrenergic receptor or insulin-like growth factor-1). Given the available evidence, we believe monoubiquitylation to be an inefficient endocytic signal in higher eukaryotes and therefore do not include it.