A single ClpS monomer is sufficient to direct the activity of the ClpA hexamer

Abstract

ClpS is an adaptor protein that interacts with ClpA and promotes degradation of proteins with N-end rule degradation motifs (N-degrons) by ClpAP while blocking degradation of substrates with other motifs. Although monomeric ClpS forms a 1:1 complex with an isolated N-domain of ClpA, only one molecule of ClpS binds with high affinity to ClpA hexamers (ClpA(6)). One or two additional molecules per hexamer bind with lower affinity. Tightly bound ClpS dissociates slowly from ClpA(6) with a t((1/2)) of approximately 3 min at 37 degrees C. Maximum activation of degradation of the N-end rule substrate, LR-GFP(Venus), occurs with a single ClpS bound per ClpA(6); one ClpS is also sufficient to inhibit degradation of proteins without N-degrons. ClpS competitively inhibits degradation of unfolded substrates that interact with ClpA N-domains and is a non-competitive inhibitor with substrates that depend on internal binding sites in ClpA. ClpS inhibition of substrate binding is dependent on the order of addition. When added first, ClpS blocks binding of both high and low affinity substrates; however, when substrates first form committed complexes with ClpA(6), ClpS cannot displace them or block their degradation by ClpP. We propose that the first molecule of ClpS binds to the N-domain and to an additional functional binding site, sterically blocking binding of non-N-end rule substrates as well as additional ClpS molecules to ClpA(6). Limiting ClpS-mediated substrate delivery to one per ClpA(6) avoids congestion at the axial channel and allows facile transfer of proteins to the unfolding and translocation apparatus.

FIGURE 1.

ClpS binding to ClpA hexamers. Binding of ClpS or ClpS-Δ17 to ClpA₆ was measured by co-migration during gel filtration (A and C) and by retention of the complex following ultrafiltration (B and D). For the co-migration assays, ClpS or ClpS-Δ17 was added in various stoichiometric ratios to 1 μm ClpA₆ in 50 mm Tris/HCl, pH 7.5, 0.1 m KCl, 10% (v/v) glycerol, 2 mm ATPγS, and 25 mm MgCl₂. The mixtures were run over a Superdex200 column in the same solution. Equal aliquots of fractions were subjected to SDS-PAGE, digital images of the Coomassie Blue-stained gels were recorded, and the distribution of ClpS or ClpS-ΔN17 was determined (see “Materials and Methods”). A, ClpS and ClpA₆ in a ratio of 1:1 (circles), 3:1 (triangles), and 6:1 (squares); fraction sizes were 80 μl each. C, ClpS-Δ17 and ClpA₆ at a ratio of 1:1 (circles), 3:1 (triangles), 6:1 (squares), and 12:1 (inverted triangles); fraction sizes were 60 μl each. For the ultrafiltration assays, ClpA was mixed with increasing amounts of [³H]ClpS (B) or [³H]ClpS-Δ17 (D) in the presence of 2 mm ATPγS and 25 mm MgCl₂ in 50 mm Tris/HCl, pH 7.5, 0.1 m KCl, and 10% (v/v) glycerol. After 5 min at room temperature, the mixture was subjected to ultrafiltration through a Microcon 100 membrane (see “Materials and Methods”). Bound [³H]ClpS or [³H]ClpS-Δ17 was calculated from the difference in free radioactive protein recovered in the filtrate and the total radioactive protein in the original mixture. The data are displayed in a Scatchard plot (bound/free versus bound), which highlights the tight binding of one ClpS and the difference in binding affinity (proportional to the slope) between ClpS and ClpS-ΔN17.

FIGURE 2.

Binding competition between ClpS and ClpSΔN17 or isolated N-domain of ClpA and slow dissociation of the complex. A, a fixed amount (0.4 μm) of [³H]ClpS was mixed with variable amounts of non-radioactive ClpS (light bars) or ClpSΔN17 (dark bars) in the presence of 0.4 μm ClpA₆ in 50 mm Tris/HCl, pH 7.5, 0.1 m KCl, 25 mm MgCl₂, 2 mm ATPγS, and 10% (v/v) glycerol. After 5 min at room temperature, bound and free [³H]ClpS were separated by ultrafiltration through Microcon 100 membranes, and bound [³H]ClpS was calculated as described in Fig. 1. B, the conditions were the same as in A except that isolated ClpA N-domain was added to compete with ClpA₆ for binding of [³H]ClpS. C, complexes of 1 eq of [³H]ClpS per ClpA hexamer (0.5 μm each) were assembled in the buffer described above. After incubation for ≥5 min at room temperature, a 30-fold excess of either non-radioactive ClpS to replace [³H]ClpS as it dissociated from ClpA₆ (open diamonds) or ClpA N-domain to trap [³H]ClpS as it dissociated from ClpA₆ (closed circles) was added. At each interval, the bound and free ClpS were separated by ultrafiltration through Microcon 100 membranes. The ClpS remaining bound to ClpA₆ was calculated as the difference between the original [³H]ClpS and the radioactivity released into the filtrate (open diamonds). D, LR-GFP_Venus degradation was initiated in the presence of ClpAP and ClpS in a standard degradation reaction mixture. Excess purified ClpA N-domain was used to trap any ClpS that dissociated from the degradative complexes, thus slowing the reaction until all ClpS had dissociated and the reaction stopped altogether. The degradation reaction was linear with time without the addition of N-domain. The dissociation of ClpS was determined from the instantaneous decrease in the degradation rate after the addition of N-domain.

FIGURE 3.

One ClpS molecule is sufficient for activation of N-end rule degradation and inhibition of non-N-end rule degradation by ClpAP. A, four series of reactions were conducted to measure degradation of LR- GFP_Venus by ClpAP in the presence of ClpS. In each series, the LR-GFP_Venus concentration was held constant at 11 μm, and the ClpS concentrations (expressed as equivalents of ClpA₆) were varied. The observed specific activities (μmol of substrate/μmol of ClpA₆/min) versus ClpS concentration are plotted. At ClpA concentrations below the K_d for dissociation with ClpS, higher amounts of ClpS are required to saturate the ClpA and achieve maximum specific activity. At 400 nm ClpA₆, only 1 eq of ClpS is needed to achieve maximum activity of ClpA₆, and higher concentrations begin to show inhibitory effects, presumably because, at free ClpS concentrations >1 μm, competition with the ClpA-ClpS complex for substrate binding becomes significant. Circles, 100 nm ClpA₆; squares, 200 nm ClpA₆; diamonds, 300 nm ClpA₆; inverted triangles, 400 nm ClpA₆. B, degradation of LR-GFP_Venus by ClpAP was measured with different ratios of ClpA and ClpS. Circles, 100 nm ClpA₆ and 100 nm ClpS; squares, 100 nm ClpA₆ and 200 nm ClpS; diamonds, 100 nm ClpA₆ and 600 nm ClpS; inverted triangles, 400 nm ClpA₆ and 400 nm ClpS. In each series, the specific activity (μmol of substrate/μmol of ClpA₆/min) is plotted as a function of substrate (LR-GFP_Venus) concentration. The apparent K_m for LR-GFP_Venus did not change, suggesting that the active species is the same with different ratios of ClpS and ClpA₆. The minimum value for K_m/k_cat is achieved with a 1:1 stoichiometry of ClpS to ClpA₆.

FIGURE 4.

Degradation of LR-GFP_Venus by ClpA hexamers with only one N-domain. Intact ClpA and ClpA-Δ153 were combined in different stoichiometric ratios and added to standard LR-GFP_Venus degradation assay solutions at a combined concentration of 0.2 μm. The solutions contained 0.5 μm ClpP and either no ClpS or 0.4 μm ClpS. A, the activities of the hybrid hexamers in the absence (light bars) and in the presence (dark bars) of ClpA plus ClpA-Δ153 are plotted for each mixture. Activity is expressed as a fraction of the specific activity of the intact ClpA hexamers. B, the ClpS-dependent activity, calculated as the difference between the assays with and without ClpS, is plotted for each molar ratio of the two proteins. The dotted line shows the fraction of hexamers with at least one N-domain; the dashed line shows the fraction with at least two N-domains. The number of intact and N-terminally deleted ClpA subunits among the hexamers in each case was calculated assuming a binomial distribution.

FIGURE 5.

One ClpS molecule is sufficient to inhibit degradation of non-N-end rule substrate by ClpAP. GFP-SsrA degradation by ClpAP was measured in standard degradation assay solutions with limiting ClpA and excess ClpP. To measure inhibition of degradation by ClpS under steady state conditions, ClpS was added together with the substrate to assays containing either 100 μm (open circles) or 400 μm ClpA₆ (open, closed, and inverted triangles). At 400 μm ClpA₆ ClpS was added either at the same time as GFP-SsrA (inverted triangles) or was preincubated with ClpA₆ for 15 (open triangles) or 30 s (closed triangles) before the addition of GFP-SsrA. Assay times were restricted to 45 s to avoid significant dissociation of ClpS, which has a slow off-rate from ClpA. The specific activity of GFP-SsrA degradation is plotted versus the molar ratio of ClpS to ClpA₆.

FIGURE 6.

ClpS displays mixed kinetics of inhibition of ClpAP activity against non-N-end rule substrates. Steady state kinetics of degradation of [³H]α-casein (A) and GFP-SsrA (B) were measured as described under “Materials and Methods.” Solutions with different concentrations of substrate protein and ClpS and all other assay components except ClpA were preincubated for 2 min at 37 °C before reactions were initiated by the addition of ClpA. Results of a single set of experiments are shown, but two other independent experiments showed the same inhibition kinetics.

FIGURE 7.

ClpS inhibition of binding of non-N-end rule substrates to ClpA. Binding of substrates to ClpA₆ was measured by co-migration of the proteins with ClpA₆ during gel filtration on a Superdex 200 column. The buffer used for assembly of the complexes and for equilibration of the column was 50 mm Tris/HCl, pH 7.5, 0.1 m KCl, 10% (v/v) glycerol, 2 mm ATPγS, and 25 mm MgCl₂. A, [³H]CI-SsrA (0.2 μm) binding to ClpA₆ (0.2 μm) with and without ClpS (1 μm). Closed triangles, [³H]CI-SsrA alone; open triangles, [³H]CI-SsrA and ClpA₆, no ClpS; closed circles, [³H]CI-SsrA incubated with ClpA₆ for 5 min prior to the addition of ClpS; open circles, ClpS incubated with ClpA₆ for 5 min prior to the addition of [³H]CI-SsrA. B, [³H]RepA (0.2 μm) binding to ClpA₆ (0.2 μm) with and without ClpS (1 μm); closed triangles, [³H]RepA alone; open triangles, [³H]RepA and ClpA₆; closed circles, [³H]RepA incubated with ClpA₆ for 5 min prior to the addition of ClpS; open circles, ClpS incubated with ClpA₆ for 5 min prior to the addition of [³H]RepA.

FIGURE 8.

Schematic drawing of the interaction between ClpS and ClpA. A molecule of ClpS (red) binds to one of the six N-domains on ClpA using the interface observed in the co-crystal of ClpS and the isolated ClpA N-domain (15, 21). Substrates with N-degrons (gold) interact with a specific binding site on ClpS and are delivered to ClpA for processing. ClpS excludes substrates lacking N-degrons (brown) by blocking access to other sites within ClpA. The flexible N-terminal portion of ClpS also contributes to binding; however, only one ClpS-N-terminal peptide can utilize this site at a time, thus restricting ClpS to one high affinity binding interaction per hexamer of ClpA. The region around residue 17 of ClpS appears to be needed for this secondary interaction, although it is not known whether it interacts on the surface of ClpA-D1 near the axial channel (upper path) or with other ClpA N-domains (lower path).