Structural basis for the allosteric activation of Lon by the heat shock protein LarA

Abstract

Lon is a conserved AAA+ (ATPases associated with diverse cellular activities) proteolytic machine that plays a key regulatory role in cells under proteotoxic stress. Lon-mediated proteolysis can be stimulated by either the unfolded or specific protein substrates accumulated under stress conditions. However, the molecular basis for this substrate-controlled proteolysis remains unclear. Here, we have found that the heat shock protein LarA, a recently discovered Lon substrate and allosteric activator, binds to the N-terminal domain (NTD) of Lon. The crystal structure of the LarA-NTD complex shows that LarA binds to a highly conserved groove in the NTD through the terminal aromatic residue of its C-terminal degron. Crystallographic and biochemical evidence further reveals that this binding exposes the hydrophobic core of LarA, which can bind a leucine residue and promote local protein unfolding. These results define the mechanistic role of the NTD in regulating Lon-mediated proteolysis in response to varying cellular conditions.

Fig. 1. LarA binds specifically to the NTD of C. crescentus Lon.

a Cartoon illustrating the composition of the protein constructs. NTD (green), LH (yellow), AAA (orange), and Protease (red) represent the N-terminal domain, long helices, ATPases Associated with diverse cellular Activities (AAA + ) domain, and protease domain of CcLon, respectively. DUF (pink) is a domain of unknown function. HTH (grey) is helix-turn-helix. b Surface representations in side (left) and top (right) views of a hexameric Lon in a substrate-free state (PDB code 7YUX). Each protomer is colored differently. One protomer is shown in ribbons. Due to the linker between the LH and AAA domains (red oval), three of the NTDs (NTD*) are rotationally flexible. c, d Chromatograms of Superose 6 size-exclusion chromatography of full-length CcLon (c) and CcLonΔ206 (d), with or without the presence of LarA. The curves of CcLon/CcLonΔ206, LarA, and protein mixtures are labeled with black, red, and blue, respectively. e Degradation of LarA and SciP by CcLon. An ATP regeneration system using ATP, creatine phosphate, and creatine kinase (CK) was included in the reactions. f No ATP-dependent degradation of LarA or SciP by CcLonΔ206. g Chromatogram of Superdex 75 size-exclusion chromatography of purified LarA-NTD complex. SDS-PAGE and Coomassie staining were used to analyze selected chromatographic fractions and the in vitro reactions. The gels and curves (c–g) shown are representative of three independent experiments with consistent results. Source data are provided as a Source Data file.

Fig. 2. Structure of LarA bound to the NTD.

a Ribbon diagrams of a complex consisting of one LarA (pink) bound to two NTD molecules (NTD: green and NTD’: gold) in the asymmetric unit (ASU) of the crystal. For clarity, the second LarA bound to NTD’ in the ASU is not shown. The amino-acid residue-binding sites identified in NTD and LarA were marked by a dotted box and a dashed box, respectively. b Close-up view of NTD bound to the C-terminal residue (His89) of LarA. The C-terminal and side-chain residues are shown in sticks. c Close-up view of the C-terminal helix of NTD and NTD’ showing the local unfolding of the latter, induced by LarA interaction (red dashed line). d Two close-up views of LarA bound to a leucine residue in the C-terminal tail of NTD’. Side chains of Val202 and Ile200 on NTD’ are labeled in grey for clear presentation.

Fig. 3. Mutational analyses of the amino-acid residue-binding pockets of the NTD and LarA.

a In vitro reactions showing the essential role of Arg29 in the NTD of CcLon for LarA-mediated degradation of SciP. b SEC-MALS analysis showing the key role of Arg29 of the NTD in mediating LarA binding. c, In vitro reactions of CcLon demonstrating the important role of Val57 in the hydrophobic binding pocket of LarA for mediating SciP degradation. The gels shown in (a) and (c) are representative of three independent experiments with consistent results. Source data are provided as a Source Data file. d In vitro reactions investigating the role of the hydrophobic residues in the C-terminal degron of SciP for LarA-mediated degradation. Densitometry analysis of SciP remaining was performed for gel data run in triplicate. p values were calculated using an unpaired two-sided Welch’s t-test. In the absence of LarA (left graph): L81D (p = 0.0012 **), L84D (p = 0.0002 ***), and I89D (p = 0.0106 *). In the presence of LarA (right graph): L81D vs. L84D + I89D (p = 0.0002 ***), L81D vs. L81D + L84D + I89D (p = 0.0015 **), L84D vs. L84D + I89D (p = 0.0017 **), L84D vs. L81D + L84D + I89D (p = 0.0098 **), I89D vs. L84D + I89D (p = 0.0012 **), and I89D vs. L81D + L84D + I89D (p = 0.0074 **). The number sign (#) in (c) and (d) denotes the various LarA or SciP constructs, which are either wild-type (wt) or with indicated mutations, used in the reactions. e Fluorescence polarization (FP) assays to detect the binding between different NTD constructs and fluorescein isothiocyanate (FITC) labeled peptides (pep-) derived from the C-terminal degron of LarA. f, g FP assays to detect the interaction of FITC-labeled peptides, derived from the wild-type (wt) C-terminal sequences of SciP (pep-SciP-wt)(f) or LarA (pep-LarA-wt)(g), with indicated proteins or protein complexes. h FP assays showing poor protein binding of the LarA or SciP peptides with all hydrophobic residues replaced by alanine (-mut). Data in (d-h) are presented as mean values ± SD; n = 3 independent measurements. Source data are provided as a Source Data file.

Fig. 4. Conformational change of LarA induced by binding to the NTD.

a Structure of the NTD-bound LarA (this work, pink) and that of free LarA (blue) predicted by AlphaFold, shown in ribbons in a similar orientation. An illustration of the secondary structure elements of the NTD-bound LarA is shown on top, with the Glu residues located in the loop regions marked. b Superimposition of the core structures of LarA in the NTD-bound and predicted free forms. c Limited proteolysis by the protease Glu-C of LarA in the presence and absence of the NTD. The LarA fragment corresponding to cleavage at the residue Glu82 is indicated by the asterisk. The gel shown is representative of two independent experiments with consistent results. Source data are provided as a Source Data file.

Fig. 5. Model for the mechanistic role of the NTD in Lon-mediated proteolysis in bacteria under various environmental conditions.

a Cartoon illustrating six NTDs distributed around the tri-tyrosine substrate-entry pore (denoted by three Ys) of hexameric Lon, drawn in top view. Each NTD harbors a binding pocket for the C-terminal aromatic or hydrophobic residues of the protein or peptide substrates. b An NTD can bind to a folded substrate, which is often overexpressed in cells under stress conditions, and induce a deformation to expose its hydrophobic core to mediate hydrophobic interactions with other substrate molecules of the same or different species, which leads to local or partially unfolding of protein substrates to facilitate binding of their C-terminal degrons to the triple tyrosine residues in the entry pore. Similar substrate-mediated hydrophobic interactions may occur by docking to the NTDs of unfolded protein substrates, which are accumulated in cells during protein unfolding stress (c) or degron peptides introduced to activate Lon-mediated proteolysis in vitro (d).