A Specificity-Enhancing Factor for the ClpXP Degradation Machine

See allHide authors and affiliations

Science  29 Sep 2000:
Vol. 289, Issue 5488, pp. 2354-2356
DOI: 10.1126/science.289.5488.2354


Events that stall bacterial protein synthesis activate the ssrA-tagging machinery, resulting in resumption of translation and addition of an 11-residue peptide to the carboxyl terminus of the nascent chain. This ssrA-encoded peptide tag marks the incomplete protein for degradation by the energy-dependent ClpXP protease. Here, a ribosome-associated protein, SspB, was found to bind specifically to ssrA-tagged proteins and to enhance recognition of these proteins by ClpXP. Cells with an sspB mutation are defective in degrading ssrA-tagged proteins, demonstrating that SspB is a specificity-enhancing factor for ClpXP that controls substrate choice.

Members of the Clp/Hsp100 adenosine triphosphatase (ATPase) family are hexameric, ring-shaped proteins that catalyze the unfolding of specific target proteins (1–8). Clp/Hsp100-catalyzed unfolding reactions have been implicated in a variety of intracellular processes, including reactivating heat-damaged proteins during stress, modulating the transformation of prionlike factors, and disassembling or degrading protein complexes involved in transposition, DNA replication, and virulence (9, 10). Many family members also participate directly in protein degradation by unfolding proteins and transporting the unfolded chain to an associated peptidase complex. For example, the ClpX unfoldase associates with the ClpP serine peptidase to form the multiring ClpXP protease (6, 11).

Clp/Hsp100 ATPases appear to recognize their substrates by binding to short, unstructured peptide sequences displayed on otherwise native proteins (12–15). The best characterized recognition peptide is the ssrA tag, AANDENYALAA, which targets proteins to the ClpX and ClpA ATPases (7,8, 12). Despite recent progress in identifying substrates for the Clp/Hsp100 proteins and the peptide signals important for their recognition, no simple sequence code has emerged that marks proteins as a specific substrate for a particular unfolding ATPase. Furthermore, although both ClpXP and ClpAP efficiently degrade ssrA-tagged proteins in vitro, ClpXP is largely responsible for degradation of these proteins in the cell (12). These observations suggested that additional cellular factors might serve to modulate substrate recognition in vivo.

Initial evidence for a ClpX-stimulatory factor was observed during purification of Escherichia coli ClpX, and a high-salt wash of partially purified ribosomes was found to be especially rich in this activity (16). Under conditions where purified ClpXP partially degraded green fluorescent protein carrying an ssrA tag (GFP-ssrA) (17), this activity stimulated degradation 10 times or more (Fig. 1A). We purified the stimulatory factor (18) until a single major protein of ∼20 kD was visible by SDS–polyacrylamide gel electrophoresis (SDS-PAGE). NH2-terminal sequencing identified this protein as SspB (stringent starvation protein B), a molecule of unknown function that is part of an operon induced by starvation (19). To demonstrate that SspB was indeed the stimulatory factor, we overproduced the protein (20) and purified it to apparent homogeneity (Fig. 1B).

Figure 1

SspB stimulates ClpXP degradation of GFP-ssrA. (A) A ribosome-associated factor stimulates proteolysis of 0.96 μM GFP-ssrA by 0.1 μM ClpX6 and 0.1 μM ClpP14. (B) MonoQ chromatography of SspB. (Inset) SDS-PAGE of fractions 14 to 17. (C) Effect of SspB on GFP-ssrA degradation. Rates were determined as described (9) in the absence (K m = 1.8 ± 0.34 μM, V max = 0.96 ± 0.06 min−1) or presence of 0.24 μM SspB (K m = 0.40 ± 0.10 μM,V max = 1.44 ± 0.07 min−1). Reactions contained 5 mM ATP, 0.3 μM ClpX6, and 0.8 μM ClpP14.

Rates of GFP-ssrA degradation by ClpXP were determined in the presence or absence of SspB (Fig. 1C). SspB reduced the Michaelis constant (K m) for this substrate by a factor of 4 to 5, indicating that it enhances productive interactions between ClpXP and ssrA-tagged proteins. SspB also stimulatedV max by about 25%. Moreover, SspB stimulated degradation over many enzyme turnovers, did not stimulate degradation of other ClpXP substrates (MuA and λ O), and did not stimulate ClpAP, which also recognizes and degrades GFP-ssrA (7). Thus, SspB enhances substrate recognition of ssrA-tagged substrates by the ClpX ATPase in a highly specific manner.

SspB bound specifically to ssrA-tagged proteins (21) (Fig. 2). SspB and GFP-ssrA coeluted from a Superose 12 column (Fig. 2A), whereas SspB and untagged GFP eluted as distinct peaks (Fig. 2B). Likewise, the ssrA-tagged NH2-terminal domain of λ repressor (λ-cI-N-ssrA) was bound by SspB (Fig. 2C). Mutagenesis of the ssrA tag revealed that residues critical for SspB binding were distinct from those recognized by ClpX. Mutations in the YALAA portion of the tag did not prevent SspB binding (Fig. 2, D to F) but severely reduced degradation by ClpXP (8, 12). By contrast, the Asn3→Ala (N3A) mutation in the AANDEN segment of the tag obliterated binding of SspB (Fig. 3A) and eliminated SspB stimulation of ClpXP degradation without affecting unstimulated degradation (Fig. 3B). Thus, Asn3 in the ssrA tag is a cardinal determinant of SspB recognition, and binding of SspB to the peptide tag is critical for stimulation of degradation. We conclude that SspB recognizes determinants in the AANDEN portion of the ssrA tag, binding adjacent to the region recognized by ClpX.

Figure 2

Gel filtration of SspB·ssrA-tagged protein mixtures. (A to F) Elution profiles of SspB (solid line) and tagged or untagged GFP or λ-cI-N substrates (dashed line). Control experiments suggest that the unbound λ-cI-N material in (C) and (D) may have lost the ssrA tag by proteolysis during purification. Abbreviations for the amino acid residues are as follows: A, Ala; D, Asp; E, Glu; L, Leu; N, Asn; and Y, Tyr.

Figure 3

Asn3 of the ssrA tag is critical for SspB binding and function. (A) SspB and GFP-ssrA-N3A chromatograph independently in gel filtration. (B) SspB does not stimulate degradation of GFP-ssrA-N3A by ClpXP. Reactions contained 5 mM ATP, 0.48 μM GFP-ssrA or GFP-ssrA-N3A, 0.02 μM ClpX6, 0.02 μM ClpP14, and 0.48 μM SspB where indicated.

SspB stimulated degradation of ssrA-tagged proteins in vivo. Pulse-chase experiments were used to determine the rate of degradation of an ssrA-tagged substrate in isogenic sspB +and sspB strains. These cells carried the wild-type clpX allele but were clpA− to allow degradation by ClpXP to be specifically measured (22). Synthesis of a tagging substrate (λ-cI-N) was induced from a gene with a strong transcriptional terminator before a stop codon; translation of this mRNA results in efficient addition of the ssrA tag to generate λ-cI-N-ssrA (23, 24). InsspB+ cells, λ-cI-N-ssrA was degraded with a half-life of ∼0.5 min (Fig. 4, A and C), whereas in sspB-defective cells its half-life was about 5 min. Over the short time-course of these experiments, no appreciable degradation was observed in the absence of ClpX in eithersspB+ or sspB cells (Fig. 4B) (25).

Figure 4

SspB stimulates intracellular degradation of ssrA-tagged proteins. (A) Pulse-chase assays of degradation of λcI-N-ssrA in sspB+ andsspB strains in cells lacking clpA. λ-cI-N-ssrA is indicated by the arrow and asterisks. (B) Experiment identical to that in (A), except that the strain wasclpA + and clpX . (C) Quantification of a pulse-chase experiment similar to that in (A).

Our results demonstrate that SspB binds to ssrA-tagged proteins and increases the efficiency with which they are recognized and degraded by ClpXP in vitro and in vivo. Moreover, the ssrA tag contains distinct sequence determinants important for recognition by SspB and by ClpX. The existence of factors, like SspB, helps explain how individual members of the Clp/Hsp100 family can efficiently recognize substrates with substantially different peptide-targeting sequences. The observation that SspB stimulates degradation by ClpXP but not ClpAP also reconciles the facts that ClpXP mediates most intracellular degradation of ssrA-tagged proteins, but both enzymes degrade these substrates with similar efficiencies in vitro (12). Overall, SspB functions as a specificity-enhancing factor in two ways: by preferentially stimulating degradation of ssrA-tagged proteins, and directing this specific class of substrates to ClpXP but not to other protease complexes.

SspB is encoded in an operon whose synthesis is stimulated by carbon, amino acid, and phosphate starvation (19), suggesting a special role during nutrient stress. Starvation is likely to increase ribosome stalling and thereby increase ssrA tagging (26). If starved cells contain higher levels of ribosome-bound SspB, these molecules could be poised to capture tagged proteins as they exit the ribosome, and this complex could then be targeted to ClpXP for degradation. Cells could therefore specifically stimulate degradation of the incomplete protein products of stalled translation, releasing amino acids for productive protein synthesis. Thus, induction of SspB would allow the turnover of individual ClpXP substrates to be differentially regulated. In contrast, inducing synthesis of ClpXP itself during starvation would lead to a global increase in degradation of all substrates, including some whose functions might be important under nutrient stress (27). Thus, recognition-enhancing specificity factors, like SspB, both help to explain how diverse substrates can be efficiently recognized by individual Clp/Hsp100 proteins and provide a way to regulate unfolding and degradation of specific protein targets in response to changing cellular environments.

  • * To whom correspondence should be addressed. E-mail: tabaker{at}


View Abstract

Stay Connected to Science

Navigate This Article