The longevity effects of reduced IGF-1 signaling depend on the stability of the mitochondrial genome

Abstract

Suppression of insulin-like growth factor-1 (IGF-1) signaling extends mammalian lifespan and protects against a range of age-related diseases. Surprisingly though, we found that reduced IGF-1 signaling fails to extend the lifespan of mitochondrial mutator mice. Accordingly, most of the longevity pathways that are normally initiated by IGF-1 suppression were either blocked or blunted in the mutator mice. These observations suggest that the pro-longevity effects of IGF-1 suppression critically depend on the integrity of the mitochondrial genome and that mitochondrial mutations may impose a hard limit on mammalian lifespan. Together, these findings deepen our understanding of the interactions between the hallmarks of aging and underscore the need for interventions that preserve the integrity of the mitochondrial genome.

Figure 1|. Lifespan, weight, size and appearance of WT, PolgA^D257A and Pappa mutant mice.

A. The lifespan of PolgA^D257A is not rescued by deletion of the Pappa gene. B-C. A homozygous deletion of the Pappa gene leads to decreased weight and increased frailty in male and female PolgA^D257A mice (male n = 8–22/group, female n = 6–13/group). D. Appearance and size distribution of WT and mutant mice (grid size = 1 cm²).

Figure 2|. The splenomegaly and anemia of PolgA^D257A mice are partially rescued by deletion of Pappa.

A-C.Male and female PolgA^D257A mice display splenomegaly. In both sexes, spleen size is reduced by deletion of two copies of the Pappa gene. In males, deletion of one copy of the Pappa gene is sufficient for spleen reduction as well. Spleen length was normalized to body length (male n = 7–17/group, female n = 6–13/group). D-E. Male and female PolgA^D257A mice exhibit reduced RBC counts, which is rescued in both sexes in PolgA^D257A; Pappa^+/− mice. F-G. Male and female PolgA^D257A mice exhibit reduced hemoglobin content, which is rescued in both sexes in PolgA^D257A; Pappa^+/− mice (male and female n = 4–8/group).

Figure 3|. Inflammation in PolgA^D257A mice is partially rescued by depletion of Pappa.

A, C, E, G. Male PolgA^D257A mice display increased levels of serum inflammation markers. IL-2 (A) and IL-6 (C) markers are rescued by deletion of one copy of the Pappa gene. B, D, F, H. Female PolgA^D257A mice only display increased levels of IL-6 among the evaluated cytokines, which is not rescued by Pappa deletion (D). (male n = 4–15/group, female n = 2–8/group).

Figure 4|. Impaired muscle and cardiac function in PolgA^D257A mice is partially rescued by deletion of Pappa.

A-D. Male and female PolgA^D257A mice display reduced grip strength and endurance. In male mice, these physiological endpoints are improved upon deletion of one Pappa copy in male, but not female mice (male n = 8–32/group, female n = 5–19/group). E-I. Male, but not female, PolgA^D257A mice display increased LVID during diastole and systole. Heart size, LVIDd and LVIDs return to WT levels in male mice upon deletion of one or two copies of the Pappa gene (male n = 7–10/group, female n = 4–9/group).

Figure 5|. Mutagenesis in WT and mutant mice.

A-C. Male PolgA^D257A mice, with or without Pappa deletion display identical mutation rates and spectra for all mutation classes. The same is observed for WT mice with or without Pappa deletion. D,E. Mutations were equally distributed across the mitochondrial genome (n = 5–6/group). First track = coverage. Second track = single nucleotide variants. Third track = insertions. Fourth track = deletions. Fifth track = mutation frequency. F. mtDNA copy number was unchanged by deletion of the Pappa gene (n = 5–6/group). G,H. Deletion of the Pappa gene reduces the number of cardiomyocytes with clonally expanded mtDNA mutations. Blue cells depict cells without WT COX activity (n = 3/group).

Figure 5|. Mutagenesis in WT and mutant mice.

A-C. Male PolgA^D257A mice, with or without Pappa deletion display identical mutation rates and spectra for all mutation classes. The same is observed for WT mice with or without Pappa deletion. D,E. Mutations were equally distributed across the mitochondrial genome (n = 5–6/group). First track = coverage. Second track = single nucleotide variants. Third track = insertions. Fourth track = deletions. Fifth track = mutation frequency. F. mtDNA copy number was unchanged by deletion of the Pappa gene (n = 5–6/group). G,H. Deletion of the Pappa gene reduces the number of cardiomyocytes with clonally expanded mtDNA mutations. Blue cells depict cells without WT COX activity (n = 3/group).

Figure 6|. Transcriptomic analysis of WT and mutant mice.

A-C. Loss of one or two copies of the Pappa gene normalizes the expression profile of PolgA^D257A mice. Genes that are upregulated (A) or downregulated (B) in PolgA^D257A mice are normalized upon deletion or depletion of Pappa. Genes that are unchanged in PolgA^D257A mice remain unchanged after complete or partial loss of Pappa (C). D. Gene expression patterns suggest that impaired cardiac and metabolic function are rescued by deletion and depletion of the Pappa gene, drastically shifting the expression patterns away from the dysregulation observed in PolgA^D257A mice. Interestingly, loss of one copy of Pappa in PolgA^D257A mice changes the transcriptional profile in the direction of Pappa^−/− compared to Pappa^+/−, suggestive of implementation of pathways to rescue the negative consequences of mtDNA mutagenesis. E. Pro-aging related gene programs are dampened by deletion and depletion of Pappa. F. Complete loss of Pappa does not restore expression patterns of mitochondrial transcripts in the PolgA^D257A background. G. Lipid metabolism related genes are not upregulated in Pappa^+/− mice or PolgA^D257A mice, but are upregulated in PolgA^D257A; Pappa^+/− mice, suggesting context dependent expression patterns (n = 6/group).

Figure 6|. Transcriptomic analysis of WT and mutant mice.

A-C. Loss of one or two copies of the Pappa gene normalizes the expression profile of PolgA^D257A mice. Genes that are upregulated (A) or downregulated (B) in PolgA^D257A mice are normalized upon deletion or depletion of Pappa. Genes that are unchanged in PolgA^D257A mice remain unchanged after complete or partial loss of Pappa (C). D. Gene expression patterns suggest that impaired cardiac and metabolic function are rescued by deletion and depletion of the Pappa gene, drastically shifting the expression patterns away from the dysregulation observed in PolgA^D257A mice. Interestingly, loss of one copy of Pappa in PolgA^D257A mice changes the transcriptional profile in the direction of Pappa^−/− compared to Pappa^+/−, suggestive of implementation of pathways to rescue the negative consequences of mtDNA mutagenesis. E. Pro-aging related gene programs are dampened by deletion and depletion of Pappa. F. Complete loss of Pappa does not restore expression patterns of mitochondrial transcripts in the PolgA^D257A background. G. Lipid metabolism related genes are not upregulated in Pappa^+/− mice or PolgA^D257A mice, but are upregulated in PolgA^D257A; Pappa^+/− mice, suggesting context dependent expression patterns (n = 6/group).