The N-end Rule (original) (raw)

1. A. Varshavsky
2. Division of Biology, California Institute of Technology, Posadena, California 91125

Excerpt

The N-end rule relates the in vivo half-life1 of a protein to the identity of its N-terminal residue (Varshavsky 1992). Similar but distinct versions of the N-end rule have been shown to operate in all organisms examined, from mammals to fungi and bacteria. I summarize the current understanding of the N-end rule pathway and describe some of the recent methods that utilize the N-end rule.

Features of a protein that confer metabolic instability are called degradation signals, or degrons (Varshavsky 1991). The essential component of one degron, the first to be identified, is a destabilizing N-terminal residue of a protein (Bachmair et al. 1986). This signal is called the N-degron. The N-end rule (defined above) results from the existence of N-degrons containing different destabilizing residues (Varshavsky 1992). In eukaryotes, the N-degron comprises two determinants: a destabilizing N-terminal residue and an internal lysine (or lysines) of a substrate. The lysine residue...

• *
  ↵* This paper is dedicated to the memory of Harold Weintraub (1945–1995)—a remarkable scientist and man of uncommon goodness.
• 1
  ↵1 In the realm of in vivo protein degradation, the term “half-life” is an approximation at best, because the corresponding decay curves deviate from an exponential (first-order) kinetics to a varying but always significant extent. This is so because a protein molecule in vivo is not a fixed structural entity. For example, the probability of degradation of a nascent, partially unfolded, chaperonin-associated protein is different from the probability of degradation of a folded counterpart of this protein at a later time in the same cell. In addition, most proteins undergo covalent modifications and associate with other molecules (including other proteins) in a cell, the fraction of a modified or a complex-associated protein being typically less than unity. Some of these modifications and associations are relevant to the relevant to the protein's function, whereas the rest of them result from a variety of quasi-random events that include protein damage. Thus, an in vivo ensemble of protein molecules encoded by one and the same ORF is inhomogeneous structurally and/or conformationally and therefore does not decay exponentially. For some short-lived proteins, e.g., Saccharomyces cerevisiae Matα2p, deviations from a first-order decay (at times comparable to a half-life) are negligible in comparison to the roughly ±5% scatter of experimental points in a standard pulse-chase protocol (Hochstrasser and Varshavsky 1990). However, many other short-lived proteins, including the engineered N-end rule substrates considered in this paper, decay with a pronounced non-first-order kinetics, the older protein molecules being longer-lived than the younger ones (Baker and Varshavsky 1991; F. Lévy et al., in prep.). A rigorous terminology that takes into account the nonexponentiality of in vivo protein degradation remains to be devised. Meanwhile, the term “half-life” is employed here as a useful approximation. A curious (but remote) analogy from another realm: The decay of radioactive elements is considered to be the paragon of a first-order process, and indeed it is, according to experimental data. Nonetheless, quantum mechanical arguments indicate that a first-order kinetics is an approximation (albeit an extremely good one) even in this case, because deviations from a first-order decay are expected to occur for times both very small and very large in comparison to a half-life (Pais 1988; Brown 1992).