The heat shock protein LarA activates the Lon protease in response to proteotoxic stress

Abstract

The Lon protease is a highly conserved protein degradation machine that has critical regulatory and protein quality control functions in cells from the three domains of life. Here, we report the discovery of a α-proteobacterial heat shock protein, LarA, that functions as a dedicated Lon regulator. We show that LarA accumulates at the onset of proteotoxic stress and allosterically activates Lon-catalysed degradation of a large group of substrates through a five amino acid sequence at its C-terminus. Further, we find that high levels of LarA cause growth inhibition in a Lon-dependent manner and that Lon-mediated degradation of LarA itself ensures low LarA levels in the absence of stress. We suggest that the temporal LarA-dependent activation of Lon helps to meet an increased proteolysis demand in response to protein unfolding stress. Our study defines a regulatory interaction of a conserved protease with a heat shock protein, serving as a paradigm of how protease activity can be tuned under changing environmental conditions.

Fig. 1. A trapping approach identifies LarA as an interactor and substrate of Lon.

a Schematic representation of the constructs used for the trapping experiment. C-terminally Twin-Strep-tagged Lon^WT or Lon^TRAP were expressed from a xylose-inducible P_xyl promoter in Δlon cells. Extracts from Δlon cells lacking Lon were used as control (No Lon). b Dot plot showing proteins enriched in the Lon^TRAP elution fraction compared to the Lon^WT and the No Lon elution fraction. Log2 values are based on averages of two independent experiments (Supplementary Dataset 1). Turquoise dots mark known/confirmed Lon substrates (FliK corresponds to CCNA_00944/45 in Supplementary Dataset 1) as well as CCNA_03707/LarA. c In vivo stability of N-terminally 3xFLAG-tagged LarA (F-LarA) in wild type (WT) and Δlon cells. Quantifications show the mean values ± SD of F-LarA levels; n = 3 or 2 (t = 45, t = 60) biologically independent samples. d Same as in (c), but with C-terminally 3xFLAG-tagged LarA (LarA-F). LarA-F levels are presented as mean values ± SD; n = 3 or 2 (t = 60) biologically independent samples. e In vitro degradation of LarA by Lon. The reactions contained 3.0 µM LarA and 0.125 µM Lon-His hexamer with (+ATP) or without (–ATP) ATP regeneration system (ATP, creatine phosphate, creatine kinase [CK]). f Quantifications showing LarA protein levels (normalised to Lon-His levels) as mean values ± SD; n = 3 independent experiments. Source data are provided as a Source Data file.

Fig. 2. LarA inhibits growth in a Lon-dependent manner.

a Growth curves of wild type C. crescentus (WT) harbouring either an empty vector (VC) or a plasmid carrying P_xyl-3xFLAG-larA (P_xyl-F-larA). Cultures were either grown under non-inducing conditions (left panel) or P_xyl-inducing conditions (+xyl; right panel). Growth curves display means ± SD; n = 9 biologically independent cultures from three independent experiments. b Same as (a) but in a Δlon background and n = 6 biologically independent cultures from two independent experiments. c Δlon cells with lon^TRAP-Twin-Strep-tag integrated on the chromosome under control of the P_xyl promoter and harbouring either an empty vector (VC) or a plasmid carrying P_xyl-3xFLAG-larA. Cultures were grown under non-inducing (left panel) or inducing conditions (+xyl). d Same as in (c) but with lon^WT-Twin-Strep-tag integrated on the chromosome under control of the P_xyl promoter. e, f Same as (c, d) but in a WT background, instead of Δlon. Growth curves in (c–f) display means ± SD; n = 3 biologically independent cultures from three independent experiments. Source data are provided as a Source Data file.

Fig. 3. LarA allosterically activates Lon-mediated degradation.

a In vivo stability of SciP and 3xFLAG-LarA (F-LarA) in WT cells carrying an empty vector (VC) or a plasmid carrying P_xyl-3xFLAG-larA (P_xyl-F-larA). Quantifications show the mean values ± SD of SciP levels; n = 3 biologically independent samples. b In vitro degradation of His-SciP (4 µM) by Lon-His hexamer (0.125 µM) in the presence (+LarA, 3 µM) or absence of LarA (–LarA). Creatine kinase [CK] was added for ATP regeneration. His-SciP protein levels (normalised to Lon-His levels) are presented as mean values ± SD; n = 3 (–LarA), 5 (+LarA) independent experiments. c Degradation rates of 2 µM His-SciP at increasing concentrations of LarA. All reactions contained 0.05 µM Lon-His hexamer. The curve represents a fitted equation considering activation and inhibition by LarA (see Methods). Values represent means ± SD; n = 3, 4 (0 µM LarA) or 5 (2 µM LarA) independent experiments. d Degradation rates at increasing His-SciP concentrations in the absence (−LarA) or presence of 2 µM LarA (+LarA). Lon-His hexamer 0.05 µM was used. Curves were fitted to Michaelis-Menten and Hill equations (see Methods). Values represent means ± SD; n = 3, 4 (2 µM − LarA) or 5 (2 µM + LarA) independent experiments. e ATPase rates of Lon-His hexamer (0.05 µM) in the presence of SciP, LarA, LarA^Δ5, either when added individually or in combination at the indicated concentrations. A reaction without substrate (−) is shown for comparison. Values represent means ± SD; n = 3, 6 (1 µM LarA), 7 (2 µM His-SciP), 12 (2 µM LarA) or 36 (−) independent reactions. f Native mass spectra of Lon^EQ alone and of Lon^EQ with either LarA or LarA^Δ5. The main charge state as well as the experimentally determined molecular weight of the detected complex is indicated. Representative data are shown; n = 3 independent experiments. Full spectra of LonEQ + LarA and LonEQ + LarA^Δ5 showing the charge states of unbound LarA and LarA^Δ5 are shown in Supplementary Fig. 4. Source data are provided as a Source Data file.

Fig. 4. LarA interacts with Lon via a C-terminal degron that is critical for Lon activation.

a Schematic representation of the C-terminal 25 amino acid (AA) residue sequence of LarA (AA 65–89) and the analysed truncation and point mutants. b, c In vivo stability of 3xFLAG-LarA (F-LarA) variants and SciP in wild type cells harbouring plasmids carrying the indicated constructs. The graphs in (c) show the mean values ± SD of F-LarA variant levels (left; n = 3 except n = 6 for F-LarA, biologically independent samples) and of SciP levels (right; n = 4 except n = 8 for VC and F-larA, biologically independent samples). d Immunoblots of 3xFLAG-tagged LarA (F-LarA) variants and Lon in lysates from WT cells (Input) and after α-FLAG immunoprecipitation. WT cells harboring either an empty vector (VC) or plasmids carrying Pxyl-3xFLAG-larA variants were grown for one hour with xylose to induce F-larA variant overexpression prior to cell lysis and IP. Lon levels after α-FLAG IP of F-LarA variants compared to the empty vector control (VC) are shown as mean values ± SD; n = 3 (DD and AA) or 5 (VC, F-larA WT and Δ5) biologically independent samples from 3 or 4 independent IP experiments, respectively. e Growth curves of WT cells harbouring an empty vector (VC) or plasmids carrying the indicated constructs, grown under non-inducing conditions (upper panel) or P_xyl-inducing conditions (+xyl; lower panel). All growth curves display means ± SD; n = 4 or 2 (Δ5) biologically independent cultures. f In vitro degradation of His-SciP in the absence (−LarA) or in the presence of the indicated LarA variants. The reactions contained 0.25 µM Lon-His hexamer and 10 µM each of His-SciP and/or the respective LarA variant. Representatives gels are shown, quantifications show means ± SD; n = 3 or 5 (+LarA^WT, −LarA) independent experiments. g In vitro degradation of the LarA variants (10 µM) by Lon-His (0.25 µM) in the presence (+ATP) or absence (−ATP) of an ATP regeneration system. Gels are representatives of three independent experiments and the quantifications show mean values ± SD; n = 3 independent experiments. Source data are provided as a Source Data file.

Fig. 5. The C-terminal LarA degron is transferable.

a Schematic representation of carboxymethylated (^CM) titin-I27 without a tag (No tag) or with the C-terminal 20, 10 or 5 amino acid residues of LarA. The titin-I27-LarA5^DD, LarA5^D and titin-I27-LarA5^AA mutants harbour the DD, D and AA substitutions, respectively, in the context of the titin-I27-LarA5 chimera. All titin-I27 variants harbour a 6xHis tag at the N-terminus. b In vitro degradation of unfolded titin-I27^CM and of the unfolded titin-I27^CM-LarA fusion constructs by Lon-His in the presence (+ATP) or absence (−ATP) of an ATP regeneration system. Quantifications show means ± SD of relative protein levels (normalised to Lon-His) from the +ATP condition; n = 3 or 4, 9 (LarA5) independent experiments. c In vitro degradation of His-SciP by Lon-His in the presence (+titin-I27^CM-LarA20) or absence (−) of titin-I27^CM-LarA20. Graph shows means ± SD of relative protein levels of His-SciP (normalised to Lon-His); n = 3 independent experiments. Source data are provided as a Source Data file.

Fig. 6. LarA enhances the degradation of a variety of Lon substrates.

a In vitro degradation of various native and artificial Lon substrates in the absence (−LarA) or presence of LarA (+LarA). Substrates were used at the following concentrations: 3 µM LarA, 8 µM β-casein, 4 µM His-SciP, FliX, FliK-C, 2 µM CcrM, 1.5 µM DnaA, 3 µM 6xHis-titinI27-β20^CM. Lon-His hexamer was added in all reactions at 0.125 µM, except for the reaction with 6xHis-titinI27-β20^CM, in which 0.075 µM Lon-His hexamer was used. Representative gels are presented, n = 3, 4 (His-SciP − LarA), 5 (His-SciP + LarA) independent experiments. b Degradation rates of the Lon substrates shown in (a) in either the absence or presence of LarA. All reactions contained 0.05 µM Lon-His hexamer, 2 µM substrate and in case of co-degradations 2 µM LarA. The fold change of the LarA-dependent effect is indicated. Quantifications present the mean values ± SD, with sample sizes as reported in (a). Statistical significance was tested using an unpaired, two-sided Welch’s t-test in R and the following p-values were obtained: His-SciP: 8.52 × 10⁻⁶ ****; CcrM: 0.008 **; FliX: 0.005 **; FliK-C: 0.022 *; DnaA: 0.439 (ns); β-casein: 0.010 *; 6xHis-titinI27-β20^CM (titin-β20^CM): 0.006 **. c ATPase rate of Lon-His (0.05 µM) in the presence of different substrates. Bars for Lon alone (−) and when incubated with either LarA or LarA^Δ5 are shown for comparison (reproduced from Fig. 3e). Labels below the bars indicate the substrate(s) and the used concentration(s). Bars and error bars represent the means ± SD;; n = 3, 4 (1 µM FliX), 5 (2 µM FliX, 2 µM FliK-C), 6 (1 µM LarA), 12 (2 µM LarA), 36 (−) independent reactions. d Immunoblots of FliK and F-LarA levels in WT cells harbouring an empty vector (VC) or plasmids carrying P_xyl-3xFLAG-larA or P_xyl-3xFLAG-larA^Δ5. Quantifications show the means ± SD of FliK levels after 0 and 60 min of xylose addition compared to the empty vector control (VC); n = 3 biologically independent samples. e Volcano plot showing proteins affected by F-LarA overexpression. WT cells harbouring either an empty vector (VC) or the plasmid carrying Pxyl-3xFLAG-larA were grown for 2 h with xylose to induce F-larA overexpression. Lon, the confirmed substrates SciP and FlgE (green label) as well as the ten most significantly changed proteins are indicated (previously reported putative substrates are labeled in purple). Analysis of three biological replicates (n = 3) using “Differential Enrichment analysis of Proteomics data“ (DEP) of bioconductor is shown (see Methods and Supplementary Dataset 2 for details and raw values). Source data are provided as a Source Data file.

Fig. 7. LarA accumulates in response to proteotoxic stress.

a Schematic representation of the operon containing CCNA_03706 (shsp2) and CCNA_03707 (larA). b Changes in shsp2, larA and lon expression induced by 6 h of DnaKJ depletion (6 h glu) compared to the non-depleted condition (xyl) in an otherwise WT strain (purple) or in a strain lacking the heat shock sigma factor σ³² (ΔrpoH) (grey). The quantifications are based on previously published RNA-sequencing data. c Immunoblots showing DnaK and LarA levels in the P_xyl-dnaKJ depletion strain and the P_xyl-dnaKJ Δlon strain. Samples were taken at the indicated time points after change of growth medium from non-depleting with xylose to DnaKJ depleting medium without xylose (−xyl). Representative data are shown; n = 2 biologically independent samples. d Immunoblots showing LarA levels in WT, Δlon and ΔlarA strains. Samples were taken before (0) and at the indicated time points after shifting the cultures from 30 °C to 45 °C. Representative data are shown; n = 3 biologically independent samples. e Immunoblot showing LarA levels in WT and Δlon strains after shifting the cultures from 30 °C to 37 °C. Representative data are shown; n = 3 biologically independent samples. f In vivo stability of LarA in WT and Δlon cells. Cultures were shifted from 30 °C to 37 °C and incubated for 25 min prior to protein synthesis shut-off at 0 min. Quantifications show LarA levels as mean values ± SD, n = 3 biologically independent samples. g Immunoblot showing induction of LarA levels in WT cells before (0 min) and after treatment (15 and 30 min) with various stress conditions inducing proteotoxic stress; heat stress at 42 °C (42 °C), addition of L-canavanine (Can; 250 μg/ml final conc.), addition of azetidine-2-carboxylate (AzC; 5 mM final), ethanol (EtOH; 5% final), kanamycin (Kan; 0.1125 μg/ml final), sodium chloride (NaCl; 85 mM final). Representative data are shown; n = 4 biologically independent samples. h Model of LarA-dependent activation of Lon at the onset of proteotoxic stress. See main text (Discussion) for a detailed description. Source data are provided as a Source Data file.