A salvage pathway maintains highly functional respiratory complex I

Abstract

Regulation of the turnover of complex I (CI), the largest mitochondrial respiratory chain complex, remains enigmatic despite huge advancement in understanding its structure and the assembly. Here, we report that the NADH-oxidizing N-module of CI is turned over at a higher rate and largely independently of the rest of the complex by mitochondrial matrix protease ClpXP, which selectively removes and degrades damaged subunits. The observed mechanism seems to be a safeguard against the accumulation of dysfunctional CI arising from the inactivation of the N-module subunits due to attrition caused by its constant activity under physiological conditions. This CI salvage pathway maintains highly functional CI through a favorable mechanism that demands much lower energetic cost than de novo synthesis and reassembly of the entire CI. Our results also identify ClpXP activity as an unforeseen target for therapeutic interventions in the large group of mitochondrial diseases characterized by the CI instability.

Fig. 1. Components of Complex I peripheral arm accumulate upon CLPP deficiency.

a Incorporation rates of CI subunits in differentiated C2C12 cells. Mitochondrial OXPHOS complexes were separated with BN-PAGE and examined by pulsed SILAC complexome analysis. Heatmap represents steady-state protein abundance (left) and assembly of newly synthesized subunits in the existing CI (right). Positions of putative ClpXP substrates (top), and visualization of exchange rates of individual CI subunits (bottom) are presented on the CI structure according to ref. . b Turnover dynamics of NDUFV2 in wild type (+/+) and CLPP-deficient (−/−) MEFs analyzed by pulse-chase immunoprecipitation. Cells were labeled with ³⁵S-methionine, proteins were isolated at indicated time points after cold-chase and immunoprecipitated with anti-NDUFV2 antibody, followed by SDS-PAGE. Relevant input fractions were used as controls. (n = 3, biologically independent experiments). c BN-PAGE followed by western blot analysis of CI in wild type (+/+) and CLPP-deficient (−/−) hearts. Fully assembled CI and N-module containing subcomplexes (asterisks) are indicated. Antibodies used are indicated in the Figure, with putative CLPP substrates shown in bold. Individual lanes represent independent biological replicates. (n = 3, biologically independent experiments). d–f Migration profiles and corresponding quantifications of N-module (d) and Q-module (e) subassemblies in wild type (+/+) and CLPP-deficient (−/−) heart mitochondria. f Relative abundance of mitochondrial respiratory complexes in heart mitochondria. Segments of complexome profiles were derived from high-resolution supramolecular complexome profiling. Bars represent mean of n = 2 biologically independent samples. g Steady-state levels of individual CI subunits and assembly factors in wild type (+/+) and CLPP-deficient (−/−) heart mitochondria. VDAC was used as a loading control. CLPP substrates are shown in bold. Bars represent mean ± SD (*p < 0.05, **p < 0.01). Unpaired Student’s t-test was used to determine the level of statistical difference (n = 3; biologically independent samples). h EPR spectra of mitochondrial membranes upon reduction with NADH in wild type (+/+) and CLPP-deficient (−/−) heart mitochondria. Signals from complex I Fe-S clusters can be identified at the low and high-field site of the predominant S3 signal originating from CII. The indicated g-values in the expanded difference spectrum mainly represent signal contributions of clusters N4, g(z) = 2.100 and g(x) = 1.885; N2, g(z) = 2.053; N3, g(x) = 1.866; several clusters are involved at g = 1.94 and 1.927 but the signal results mainly from N1b g(x,y) and N2 g(x,y). Spectra were normalized to the same protein concentration; ten scans were accumulated for each spectrum.

Fig. 2. ClpXP protease regulates the turnover of N-module.

a Time-lapse BN-PAGE followed by western blot analysis of CI profiles in wild type (+/+) and CLPP-deficient (−/−) MEFs upon acute respiratory chain disruption via doxycycline-mediated inhibition of mitochondrial protein synthesis (n = 3). b BN-PAGE followed by western blot analysis of CI in wild type (+/+) and CLPP-deficient (−/−) cells upon CRISPR/Cas9 depletion of NDUFB11 subunit (n = 3). c Import of radiolabeled precursors (pre-) of CI subunits into wild type (+/+) and CLPP-deficient (−/−) mitochondria isolated from MEFs (n = 4). d Import of radiolabelled NDUFV2 and NDUFS3 precursors and subsequent incorporation into Complex I in intact heart mitochondria from wild type (+/+) and CLPP-deficient (−/−) animals. After the indicated incubation times mitochondrial complexes were analyzed by BN-PAGE (n = 4). e Pulse labeling (³⁵S-methionine) of mtDNA-encoded subunits in wild type (+/+) and CLPP-deficient (−/−) MEFs followed by chases for indicated time points. Mitochondrial complexes were analyzed by BN-PAGE (n = 3). Antibodies used were raised against proteins indicated in the Figure, with putative CLPP substrates shown in bold. “n” represents number of biologically independent experiments.

Fig. 3. In the absence of CLPP N-module is not exchanged on CI.

a De novo synthesis of mitochondrial proteins in wild type (+/+) and CLPP-deficient (−/−) MEFs followed by SDS-PAGE. Coomassie blue staining was used as a loading control. Individual lanes represent technical replicates of the wild type and CLPP-deficient MEF lines characterized by similar rates of mitochondrial translation. b Mitochondrial oxygen consumption rates (OCR; pmol O₂/min) in wild type (+/+) and CLPP-deficient (−/−) MEFs and HeLa cells. Bars represent means ± SD. (MEFs n = 5–6; HeLa n = 6; biologically independent samples). Specific inhibitors used in the analysis are indicated. c Turnover of the CI subunits in wild type (+/+) and CLPP-deficient (−/−) MEFs. Cells were examined using pulsed SILAC complexome analysis after separation with BN-PAGE. Heatmap (top) shows protein abundance of previously synthesized proteins (light amino acids). Heatmap (bottom) shows newly biosynthesis and assembly of subunits in existing stable CI. Right panel represents the exchange rates of individual CI subunits on a fully assembled complex visualized on the CI structures. d, e Exchange rates of CI subunits in wild type (+/+) and CLPP-deficient (−/−) MEFs. Cells were examined using pulsed SILAC complexome analysis after separation with BN-PAGE. Bars represent mean ± SD. (**p < 0.01, ***p < 0.001). Paired Student’s t-test was used to determine the level of statistical difference. Detailed analysis is described in Supplementary Data 5. d Exchange rates of either all CI subunits (Assembl. CI) or only three (NDUFS1, NDUFV1, and NDUFV2) core N-module subunits (N-core) at the level of fully assembled CI (fractions 7628-995, n = 14) were calculated as ratio of newly synthesized to steady-state levels. e (left) Newly synthesized (new) and steady-state levels of core N-module subunits (NDUFS1, NDUFV1, and NDUFV2) at the level of free N-module subcomplexes (fractions 69-7, +/+ n = 12; −/− n = 15); (right) Exchange rates of core N-module subunits calculated as ratio of newly synthesized (red) to steady state (black) levels.

Fig. 4. Accumulation of modified N-module causes respiratory chain deficiency.

a Upper panel: BN-PAGE followed by western blot analysis of CI profiles upon the siRNA-mediated Clpx knockdown in wild type (+/+) and CLPP-deficient (−/−) MEFs. Lower panel: western blot analysis of CLPP and CPLX levels upon siRNA-mediated knockdown of Clpx. (n = 4). b–d Inverse-shift analysis of CI subunits in wild type (+/+) or CLPP-deficient (−/−) MEFs investigated by western blot. The NEM-mediated stabilization of the free thiols and the subsequent mmPEG₂₄-modification after reductant treatment (TCEP) of previously blocked thiols allowed the mass-shift differentiation within the protein pool. (min = free thiols; max = bound thiols). b MEFs were pre-incubated in the absence or presence of ROS-inducing rotenone (ROT; 200 μM) or paraquat (PQ; 1 mM). Each genotype is represented by two independent clones (n = 3); c Analysis performed upon 48 h siRNA-mediated knockdown of Lonp1 or Clpx (n = 2); d Upper panel: inverse-shift analysis of the individual CI subunits in wild type (WT), CLPP-deficient (CLPP KO), NDUFB11-deficient (B11 KO), and CLPP/NDUFB11 double-deficient (DKO) MEFs. Lower panel: schematic representation of the experimental outcome. e Native-immunoprecipitation efficiency of NDUFV2 subunit from wild type (+/+) and CLPP-deficient (−/−) heart mitochondria using various solubilization conditions: NaCl–salt extraction; digitonin–lysis in 1% digitonin (w/v); Triton-X-100–lysis in 1% Triton-X-100 (w/v) (n = 2). a–e Antibodies used were raised against proteins indicated in the Figure, with putative CLPP substrates shown in bold. “n” represents number of biologically independent experiments. Individual lanes represent biological replicates. f FMN content in intact mitochondria and permeabilized mitochondrial membranes from wild type (WT), CLPP-deficient (CLPP KO), NDUFB11-deficient (B11 KO), and CLPP/NDUFB11 double-deficient (DKO) MEFs. Cells were incubated for 16 h in the presence or absence of mitoPQ (5 μM) prior to mitochondria isolation. Mitochondria from different experiments (n = 3–6) were pooled and further preceded as technical replicates (n = 6). Flavin content was determined with fluorescence record (at excitation/emission 470/525 nm), and calculated according to calibration with FMN standards. Bars represent mean ± SD (***p < 0.001). Two-way ANOVA followed by Tukey’s multiple comparisons post-test was used to determine the level of statistical difference.

Fig. 5. The N-module turnover protects from increased oxidative stress.

a, b BN-PAGE followed by western blot analysis of CI in wild type (+/+) and CLPP-deficient (−/−) MEFs treated with rotenone (ROT; 200 μM), N-acetylcysteine (NAC; 2 mM) and β-nicotinamide adenine dinucleotide hydrate (NAD; 2 mM) or antimycin A (25 μM). (n = 3, biologically independent experiments). c Survival of wild type (+/+) and CLPP-deficient (−/−) MEF cells upon 16 h treatment with rotenone (ROT; 200 μM) ± NAC (2 mM). Values are calculated as a percentage of the control (CTRL) cells. Bars represent mean ± SD. (*p < 0.05). Two-way ANOVA followed by Sidak’s multiple comparison post-test was used to determine the level of statistical difference. (CTRL & NAC n = 6; ROT & ROT + NAC n = 10, biologically independent samples). d BN-PAGE followed by western blot analysis of CI profiles in wild type (+/+) and CLPP-deficient (−/−) MEFs treated with potassium cyanide (KCN; 1 mM) and myxothiazol (1 μM) for 16 h. (n = 2, biologically independent experiments). a, b, d Antibodies were raised against proteins indicated in the panels, with putative CLPP substrates shown in bold. CI depleted of N-module are indicated with the arrowheads. e Quantification of oxidized proteins in wild type and CLPP-deficient heart mitochondria by mass spectrometry. Values were normalized to mean protein abundance. Data are shown in volcano plots using permutation-based FDR calculation (FDR < 0.05), (n = 3, biologically independent experiments). Red color represents significantly more oxidized proteins in Clpp−/− mitochondria, and blue in Clpp+/+ samples. CLPP targets are indicated in bold. f Ratiometric imaging for response to hydrogen peroxide of Hela wild type and CLPP-deficient cells. Cells containing the matrix-targeted H₂O₂ sensor HyPer3 or as pH-control matrix-targeted SypHer sensors were analyzed for their response towards 45 µM hydrogen peroxide. The SypHer sensor was not deflected during the experiment indicating that the HyPer3 response was solely due to changes in hydrogen peroxide levels. Results were from two independent experiments (n[HeLa, HyPer3] = 141 cells, n[ClpP KO, HyPer3] = 70 cells, n[HeLa, SypHer] = 89 cells, n[ClpP KO, SypHer] = 75 cells. Shapiro–Wilk test to test for normal distribution followed by Mann–Whitney-U-test was used to determine the level of statistical difference (***p < 0.001).

Fig. 6. Loss of CLPXP ameliorates respiratory chain deficiency caused by OXPHOS complex instability.

a BN-PAGE followed by western blot analysis of CI in wild type (WT) and PolgA^mut/PolgA^mut (mtDNA mutator) MEFs upon the 48 h siRNA-mediated Clpx knockdown. SDHA and VDAC were used as a loading control (n = 2, biologically independent experiments). b Proliferation capacity of mtDNA mutator MEFs upon siRNA-mediated Clpx knockdown (Clpx siRNA) in comparison to control (control siRNA) 48 h after transfection. Bars represent mean ± SD. (**p < 0.01). Unpaired Student’s t-test was used to determin the level of statistical difference (n = 6, biologically independent samples). c BN-PAGE followed by western blot analysis of CI in wild type (WT) and CIV-deficient COX10 knockout (Cox10) fibroblasts upon the 48 h siRNA-mediated Clpx knockdown. Bottom panels show the abundance of fully assembled Complex II, III, and IV in relevant samples (n = 2, biologically independent experiments). d CLPP downregulation ameliorates the detrimental phenotypes associated with CI deficiency in roundworm C. elegans. Wild type (N2) and gas-1(fc21) mutant worms were exposed to the Clpp-targeting RNAi. The developmental progress was assayed as a proportion of worms that reached a particular stage four days after hatching (left panel). The brood size was estimated as a number of living progeny per adult worm. Wild-type worms were used as control. Bars represent mean ± SD (*p < 0.05, **p < 0.01, ***p < 0.001). For developmental assay, statistical analysis was performed using Fisher Exact Test (n = 4, independent biological samples with 100 animals per condition). For brood size assay, data was compared using paired Student’s t-test. Larval stage 3 (L3), larval stage 4 (L4), adulthood (D1) (n = 10, animals per condition). e BN-PAGE followed by western blot analysis of CI profiles in a wild type (N2) and CI-deficient (gas-1) worms upon the RNAi-mediated downregulation of CLPP protease. a–e Antibodies used were raised against proteins indicated in the Figure, with putative CLPP substrates shown in bold. SDHA, VDAC1, and ATP5A were used as a loading control. f A comprehensive model for the ClpXP-mediated CI surveillance in wild type condition, as well as in proliferating or postmitotic (heart) cells.