A proteolytic AAA+ machine poised to unfold protein substrates

Abstract

AAA+ proteolytic machines unfold proteins before degrading them. Here, we present cryoEM structures of ClpXP-substrate complexes that reveal a postulated but heretofore unseen intermediate in substrate unfolding/degradation. A ClpX hexamer draws natively folded substrates tightly against its axial channel via interactions with a fused C-terminal degron tail and ClpX-RKH loops that flexibly conform to the globular substrate. The specific ClpX-substrate contacts observed vary depending on the substrate degron and affinity tags, helping to explain ClpXP's ability to unfold/degrade a wide array of different cellular substrates. Some ClpX contacts with native substrates are enabled by upward movement of the seam subunit in the AAA+ spiral, a motion coupled to a rearrangement of contacts between the ClpX unfoldase and ClpP peptidase. Our structures additionally highlight ClpX's ability to translocate a diverse array of substrate topologies, including the co-translocation of two polypeptide chains.

Fig. 1. ClpXP bound to degron-tagged DHFR.

A Cartoon of steps in ClpXP degradation of a protein substrate. B Overlay of cryoEM map and model of ClpXP bound to the branched-degron-DHFR substrate (left). Density for the degron tail could be modeled extending from the folded domain of DHFR to the ClpP entry portal (center), with the maleimide-cysteine branchpoint marked by a yellow sphere. The insets (right) display density for methotrexate and the DHFR Trp30 and Phe31 side chains (top), and the last two residues of native DHFR, Arg¹⁵⁸, and Arg¹⁵⁹, which were contacted by the pore-1 loop of ClpX subunit A, which includes Tyr¹⁵³ (bottom).

Fig. 2. Fully engaged DHFR can assume distinct orientations with respect to ClpX₆ and its RKH loops.

ClpX is shown in outline representation with the positions of subunits A¹, B², C³, and D⁴ marked and the RKH loops of these subunits shown as spheres in different colors. The DHFR substrate is depicted in cartoon/outline representation in a rainbow-color scheme, with blue representing the N-terminus and red the C-terminus. A DHFR•MTX positioning in the branched-degron structure. B DHFR•MTX positioning in the linear-degron structure.

Fig. 3. Interactions between ClpX axial pore-loops and DHFR-degron tails.

Pore-1 loops (residues 150-155) are shown as spheres, pore-2 loops (residue 198-205) are shown as cartoons, and DHFR is shown in red surface representation. A Front and back views for the branched-degron substrate. B Front and back views for the linear degron substrate. Note that in each structure, the pore-1 loop from the F^seam subunit is not engaged with the substrate.

Fig. 4. Rearrangement of ClpX-ClpP contacts correlates with upward movement of the seam subunit.

In many ClpXP complexes, the empty binding cleft on a ClpP₇ ring lies between the clefts occupied by the IGF loops of ClpX subunits E⁵ and F^seam (A; PDB code 6WRF). In the structures presented here and some other ClpXP structures (B; PDB code 8V9R), the IGF loop of chain E⁵ moves into a binding cleft that is unoccupied in structures similar to that in (A). These movements correlate with F^seam upward movement, as shown in (C, D).

Fig. 5. Rates of ATP hydrolysis by ClpXP in the absence and presence of a degron-tagged DHFR•MTX substrate.

Bars show mean rates of ATP hydrolysis for ClpXP, or ClpXP in the presence of branched-degron tagged DHFR•MTX ± 1 SD, with symbols denoting replicate (n = 3) assays.