Cdc48 (p97): a ‘molecular gearbox’ in the ubiquitin pathway?

Cdc48 (p97), a conserved chaperone-like ATPase of eukaryotic cells, has attracted attention recently because of its wide range of cellular functions. Cdc48 is intimately linked to the ubiquitin pathway because its primary action is to segregate ubiquitinated substrates from unmodified partners. This ‘segregase’ activity is crucial for certain proteasomal degradation pathways and for some nonproteolytic functions of ubiquitin. Cdc48 associates not only with different ‘substrate-recruiting cofactors’ but also with distinct ‘substrate-processing cofactors’. The latter proteins control the degree of ubiquitination of bound substrates by shifting the polyubiquitination reaction into ‘forward’, ‘neutral’ or ‘reverse’. We discuss how Cdc48 might use this ‘gearbox activity’ to control protein fate and propose a similar mode of action for the 19S cap of the proteasome.

Introduction

The conserved homohexameric ring-shaped AAA (ATPase associated with various activities) ATPase, called Cdc48 in budding yeast and p97 (the unfavorable name VCP, for valosin-containing protein was given after an artefact) in mammals, is a central component of the ubiquitin system. As the yeast name indicates, Cdc48 was initially identified as a protein required for progression through the cell division cycle. Studies in yeast also provided the first link between Cdc48 and the ubiquitin pathway, with the finding that the enzyme is required for the degradation of some artificial model substrates (linear ubiquitin–protein fusions) [1]. Later work showed that Cdc48 (p97) is involved in ubiquitin-dependent activation of certain transcription factors 2, 3, 4, the degradation of proteins of the endoplasmic reticulum (ER) by the ER-associated degradation (ERAD) pathway 5, 6, 7, 8, 9, and the control of membrane fusion 10, 11, 12, 13, 14. Whether Cdc48 (p97) also functions outside the ubiquitin system is unknown but seems unlikely. Most if not all of the known Cdc48 (p97)-dependent functions seem to be directly linked to the ability of the protein to bind to (oligo)ubiquitinated proteins and to segregate them from their binding partners, or to extract them from protein complexes 3, 15. This ‘segregase’ function is mediated by the Cdc48 (p97) ATPase activity, which translates ATP hydrolysis into mechanical forces that move and partially rotate the outside rim of the ring-shaped enzyme [16]. Cdc48 (p97) possesses two consecutive AAA ATPase domains (called D1 and D2) and an N-terminal domain (N-domain). How Cdc48 (p97) associates with substrates is an area of active research, and two possible mechanisms have been found. First, it might bind to ubiquitinated substrates directly by its N-domain, as indicated by in vitro binding studies 3, 17. Second, it might bind to ubiquitinated substrates indirectly through cofactors 3, 15, 18, 19, 20. Indeed, the second mechanism might be more common, because numerous putative Cdc48 (p97) ‘substrate-recruiting cofactors’ have been identified recently, which possess ubiquitin-binding domains and usually interact with Cdc48 (p97) by its N-domain (Box 1 and Table 1).

Remarkably, Cdc48 (p97) functions not only as a segregase; recent findings indicate that it also controls the degree of ubiquitination of the bound substrates 18, 21. This activity is brought about by so-called ‘substrate-processing cofactors’ of Cdc48 (p97) that either promote polyubiquitination, inhibit polyubiquitination or even deubiquitinate the bound (oligo)ubiquitinated substrate 13, 18, 21, 22, 23, 24 (Box 1 and Table 1).

The diverse functions, structure and mechanistic details of the Cdc48 (p97) enzyme have been excellently reviewed previously 25, 26, 27. Instead, here we discuss when and how Cdc48 (p97) is employed in ubiquitin-dependent pathways. Also, for reasons of simplicity, we will focus mainly on the yeast proteins and processes involved, although the principles discussed apply equally to the metazoan p97 protein (unless otherwise indicated). Specifically, we propose that Cdc48 (p97) might function similarly to a gearbox in a car and might control protein fate. We will speculate about the potential usefulness of such a ‘gearbox’ activity within the ubiquitin pathway and argue that it might be crucial for shifting the system from nonproteolytic to proteolytic functions of the ubiquitin system. Finally, we speculate that the 19S cap of the proteasome might have a similar mode of action.