Autophagy compensates impaired energy metabolism in CLPXP-deficient Podospora anserina strains and extends healthspan

Abstract

The degradation of nonfunctional mitochondrial proteins is of fundamental relevance for maintenance of cellular homeostasis. The heteromeric CLPXP protein complex in the mitochondrial matrix is part of this process. In the fungal aging model Podospora anserina, ablation of CLPXP leads to an increase in healthy lifespan. Here, we report that this counterintuitive increase depends on a functional autophagy machinery. In PaClpXP mutants, autophagy is involved in energy conservation and the compensation of impairments in respiration. Strikingly, despite the impact on mitochondrial function, it is not mitophagy but general autophagy that is constitutively induced and required for longevity. In contrast, in another long-lived mutant ablated for the mitochondrial PaIAP protease, autophagy is neither induced nor required for lifespan extension. Our data provide novel mechanistic insights into the capacity of different forms of autophagy to compensate impairments of specific components of the complex mitochondrial quality control network and about the biological role of mitochondrial CLPXP in the control of cellular energy metabolism.

Keywords: Podospora anserina; CLPXP protease; aging; autophagy; energy metabolism; mitochondria.

Figure 1

Mitochondrial function is partially impaired in ∆PaClpXP compared to wild type. (a) Relative oxygen consumption rate (OCR) of 7‐day‐old ∆PaClpXP and wild type mitochondria (for each strain, three mitochondrial preparations with three to six technical replicates were analyzed). State 4 OCR of the wild type was set to 100%. (b) The relative oxygen consumption of mycelia from 7‐day‐old wild type and ∆PaClpXP strains (three biological and four technical replicates) was analyzed after KCN (inhibition of C IV) or SHAM [inhibition of alternative oxidase (AOX)] addition. The basal oxygen consumption of the wild type was set to 100%, and the inhibition of the oxidases was calculated relatively. (c) Basal oxygen consumption rate (OCR) per mg dry weight from mycelia of 7‐day‐old wild type and ∆PaClpXP strains (three biological and four technical replicates). (d) BN‐PAGE analysis of 100 μg mitochondrial protein extracts from 7‐day‐old wild type and ∆PaClpXP (three mitochondrial preparations). SC, supercomplexes, V₂, complex V dimer; I₁, complex I monomer; V₁, complex V monomer; III ₂, complex III dimer; IV ₁, complex IV monomer. Changes in the composition are marked with arrows. Coomassie‐stained gel (left) and ‘in‐gel’ staining of C V (right) are demonstrated. (e) Quantification of relative complex I (CI, Coomassie‐stained gel) and complex V (CV, CV‐stained gel) protein levels of wild type and ∆PaClpXP mitochondrial extracts (n = 3) that were separated during a BN‐PAGE analysis shown in (d). Protein abundance in the wild type was set to 1. Error bars correspond to the standard deviation. P‐values were determined by Student's t‐test (P < 0.05). (f) ATP levels of 7‐day‐old wild type vs. ∆PaClpXP (three biological replicates with three technical replicates) were measured by a luminescence‐based assay. Error bars correspond to the standard deviation, and P‐values were determined with paired Student's t‐test (P > 0.05). (g) Determination of superoxide and hydrogen peroxide in 7‐day‐old wild type and ∆PaClpXP strains by NBT and DAB staining. (a–c) Error bars correspond to the standard deviation, and P‐values were determined by two‐tailed Mann–Whitney–Wilcoxon U‐test. (a, b, e) *= P< 0.05, **= P< 0.01, *** P< 0.001.

Figure 2

It is healthy to preserve mitochondria from degradation by mitophagy in the different PaClpXP deletion strains. (a) Monitoring mitophagy by western blot analysis of 7‐ and 20‐day‐old PaSod3::Gfp compared to ∆PaClpX/PaSod3::Gfp, ∆PaClpP/PaSod3::Gfp, and ∆PaClpXP/PaSod3::Gfp. (b) Quantification of ‘free GFP’ protein levels of 7 (n = 3)‐ and 20 (n = 4)‐day‐old PaSod3::Gfp vs. ∆PaClpX/PaSod3::Gfp, ∆PaClpP/PaSod3::Gfp, and ∆PaClpXP/PaSod3::Gfp normalized to the level of PaSOD1. Protein abundance in 7‐day‐old PaSod3::Gfp was set to 1. Error bars correspond to the standard deviation. P‐values were determined by Student's t‐test. (c) Relative growth rates of MMS‐treated wild type (n = 16) and ∆PaClpXP (n = 17) compared to the untreated controls. P‐values were determined between the wild type and the mutant and between each strain and MMS condition by two‐tailed Mann–Whitney–Wilcoxon U‐test. Error bars correspond to the standard deviation. (d) Survival curves of the wild type (n = 20), ∆PaClpXP (n = 21), ∆PaCypD (n = 23), and ∆PaClpXP/∆PaCypD (n = 23). P‐values were determined between wild type and ∆PaClpXP (P < 0.001), wild type and ∆PaCypD (P > 0.05), ∆PaClpXP and ∆PaCypD (P < 0.001), between ∆PaClpXP/∆PaCypD and ∆PaClpXP (P < 0.05), ∆PaClpXP/∆PaCypD and wild type (P < 0.001), respectively, ∆PaClpXP/∆PaCypD and ∆PaCypD (P < 0.001) by two‐tailed Mann–Whitney U‐test. (e) Growth rate of the wild type (n = 20), ∆PaClpXP (n = 22), ∆PaCypD (n = 24), and ∆PaClpXP/∆PaCypD (n = 26). Error bars correspond to the standard deviation, and P‐values were determined by two‐tailed Mann–Whitney–Wilcoxon U‐test. (f) Monitoring mitophagy by western blot analysis of PaSod3::Gfp compared to ∆PaClpXP/PaSod3::Gfp, ∆PaCypD/PaSod3::Gfp, and ∆PaClpXP/∆PaCypD/PaSod3::Gfp. The Coomassie‐stained gel serves as a loading control. (g) Determination of superoxide and hydrogen peroxide in 7‐day‐old wild type, ∆PaClpXP, ∆PaCypD, and ∆PaClpXP/∆PaCypD strains by NBT and DAB staining. (c, e) *= P< 0.05, **= P< 0.01, *** P< 0.001.

Figure 3

Autophagy is induced in a starvation‐like response of PaClpXP mutant strains. LSFM of hyphae from 4 (a)‐ and 20 (b)‐day‐old wild type and ∆PaClpX, ∆PaClpP, and ∆PaClpXP strains expressing Gfp::PaAtg8. (c) Quantification of autophagosomes of 4‐ and 20‐day‐old wild type and ∆PaClpX, ∆PaClpP, and ∆PaClpXP strains expressing Gfp::PaAtg8 (n = 10). P‐values were determined between 4‐ and 20‐day‐old strains and between wild type and mutant of the same age. Error bars correspond to the standard error. P‐values were determined by two‐tailed Mann–Whitney–Wilcoxon U‐test. (d) Monitoring autophagy by western blot analysis of 7‐ and 20–day‐old PaSod1::Gfp compared to ∆PaClpX/PaSod1::Gfp, ∆PaClpP/PaSod1::Gfp, and ∆PaClpXP/PaSod1::Gfp cultivated on CM medium. (e) Monitoring autophagy by western blot analysis of 7‐day‐old PaSod1::Gfp compared to ∆PaClpX/PaSod1::Gfp, ∆PaClpP/PaSod1::Gfp, and ∆PaClpXP/PaSod1::Gfp cultivated on M2 medium. (f) Quantification of ‘free GFP’ protein levels of 7 (n = 4)‐day‐old PaSod1::Gfp vs. ∆PaClpX/PaSod1::Gfp, ∆PaClpP/PaSod1::Gfp, and ∆PaClpXP/PaSod1::Gfp cultivated on CM or M2 medium normalized to the level of PaSOD1. Protein abundance in 7‐day‐old PaSod1::Gfp was set to 1. Error bars correspond to the standard deviation. P‐values were determined by Student's t‐test. (g) Survival curves of the wild type (n = 25) and ∆PaClpXP (n = 25) cultivated on M2 vs. M2‐. P‐values were determined between wild type and ∆PaClpXP on M2 (P < 0.001), wild type and ∆PaClpXP on M2‐ (P < 0.001), wild type on M2 and M2‐ (P < 0.001), and ∆PaClpXP on M2 and M2‐ (P < 0.001) by two‐tailed Mann–Whitney–Wilcoxon U‐test. (h) ATP levels of 5‐day‐old wild type, ∆PaClpXP, and ∆PaClpXP/∆PaAtg1 (three biological replicates with three technical replicates) were measured by a luminescence‐based assay. Error bars correspond to the standard deviation, and P‐values were determined with paired Student's t‐test. (c, f, h) *= P< 0.05, **= P< 0.01, *** P< 0.001.

Figure 4

Autophagy is responsible for the healthy phenotype of ∆PaClpXP mutants. (a) Survival curves of the wild type (n = 27), ∆PaAtg1 (n = 27; P < 0.001), ∆PaClpX (n = 29), and ∆PaClpX/∆PaAtg1 (n = 30; P < 0.001). (b) Relative mean lifespan of ∆PaAtg1 (n = 27), ∆PaClpX (n = 29), and ∆PaClpX/∆PaAtg1 (n = 30) resulting from the comparison of the mean lifespan of each strain with the mean lifespan of the wild type (n = 27, set to 100%). (c) Relative mean growth rates of ∆PaAtg1 (n = 27), ∆PaClpX (n = 29), and ∆PaClpX/∆PaAtg1 (n = 30) derived from the comparison of the mean growth rate of each strain with the mean growth rate of the wild type (n = 27, set to 100%). (d) Survival curves of the wild type (n = 27), ∆PaAtg1 (n = 27; P < 0.001), ∆PaClpP (n = 32), and ∆PaClpP/∆PaAtg1 (n = 23; P < 0.001). (e) Relative mean lifespan of ∆PaAtg1 (n = 27), ∆PaClpP (n = 32), and ∆PaClpP/∆PaAtg1 (n = 23) resulting from the comparison of the mean lifespan of each strain with the mean lifespan of the wild type (n = 27, set to 100%). (f) Relative mean growth rates of ∆PaAtg1 (n = 27), ∆PaClpP (n = 32), and ∆PaClpP/∆PaAtg1 (n = 23) derived from the comparison of the mean growth rate of each strain with the mean growth rate of the wild type (n = 27, set to 100%). (g) Survival curves of the wild type (n = 27), ∆PaAtg1 (n = 27; P < 0.001), ∆PaClpXP (n = 31), and ∆PaClpXP/∆PaAtg1 (n = 25; P < 0.001). (h) Relative mean lifespan of ∆PaAtg1 (n = 27), ∆PaClpXP (n = 31), and ∆PaClpXP/∆PaAtg1 (n = 25) resulting from the comparison of the mean lifespan of each strain with the mean lifespan of the wild type (n = 27, set to 100%). (i) Relative mean growth rates of ∆PaAtg1 (n = 27), ∆PaClpXP (n = 31), and ∆PaClpXP/∆PaAtg1 (n = 25) derived from the comparison of the mean growth rate of each strain with the mean growth rate of the wild type (n = 27, set to 100%). (b, c, e, f, h, i) Error bars correspond to the standard error, and P‐values were determined by two‐tailed Mann–Whitney–Wilcoxon U‐test. *= P< 0.05, **= P< 0.01, *** P< 0.001.

Figure 5

The healthy phenotype of ∆PaIap is not autophagy dependent. (a). LSFM of hyphae from 4‐ and 20–day‐old wild type and ∆PaIap strains expressing Gfp::PaAtg8. (b) Quantification of autophagosomes of 4‐ and 20‐day‐old wild type and ∆PaIap strains expressing Gfp::PaAtg8 (n = 10). P‐values were determined between 4‐ and 20‐day‐old strains and between wild type and mutant of the same age. Error bars correspond to the standard error. (c) Survival curves of the wild type (n = 27), ∆PaAtg1 (n = 27; P < 0.001), ∆PaIap (n = 46), and ∆PaIap/∆PaAtg1 (n = 38; P < 0.01). (d) Relative mean lifespan of ∆PaAtg1 (n = 27), ∆PaIap (n = 46), and ∆PaIap/∆PaAtg1 (n = 38) resulting from the comparison of the mean lifespan of each strain with the mean lifespan of the wild type (n = 27, set to 100%). (e) Relative mean growth rates of ∆PaAtg1 (n = 27), ∆PaIap (n = 46), and ∆PaIap/∆PaAtg1 (n = 38) derived from the comparison of the mean growth rate of each strain with the mean growth rate of the wild type (n = 27, set to 100%). (f) Monitoring mitophagy during MMS treatment (0.09% for the last 5 h of cultivation) by western blot analysis of 7‐day‐old PaSod1::Gfp compared to ∆PaIap/PaSod1::Gfp. (g) Monitoring autophagy by western blot analysis of 7‐day, 10‐day, 20‐day, and 24‐day‐old PaSod1::Gfp compared to ∆PaIap/PaSod1::Gfp. (h) Quantification of ‘free GFP’ protein levels of 7‐day (n = 4), 10‐day (n = 4), 20‐day (n = 4), and 24‐day (n = 4)‐old PaSod1::Gfp vs. ∆PaIap/PaSod1::Gfp normalized to the level of PaSOD1. Protein abundance in 7–day‐old PaSod1::Gfp was set to 1. Error bars correspond to the standard deviation. P‐values were determined by Student's t‐test. (b, d, e) P‐values were determined by two‐tailed Mann‐Whitney‐Wilcoxon U‐test (*= P< 0.05, **= P< 0.01, *** P< 0.001).

Figure 6

Regulation of aerobic metabolism and autophagy in the wild type compared to ∆PaClpXP. In the wild type, NADH generated by the tricarboxylic acid cycle (TCA) is used by oxidative phosphorylation to generate ATP, and thus only basal autophagy levels occur. Because there are no limitations in mitochondrial function, mitophagy level and lifespan behave normal in Podospora anserina wild type. Ablation of the mitochondrial AAA+ protease PaCLPXP leads to modifications in mitochondrial respiratory (complex I, complex IV, and complex V) chain and energy metabolism. This deficiency is compensated by molecular responses leading to the induction of the alternative oxidase (AOX) and by the upregulation of unspecific general autophagy, compensating for a decreased ATP content and leading to an increased lifespan of PaClpXP deletion mutants as a hormetic response to mild stress. Figure is modified according to Rizzuto et al. (2012). OMM, outer mitochondrial membrane; IMM, inner mitochondrial membrane.