Longitudinal autophagy profiling of the mammalian brain reveals sustained mitophagy throughout healthy aging

Abstract

Mitophagy neutralizes mitochondrial damage, thereby preventing cellular dysfunction and apoptosis. Defects in mitophagy have been strongly implicated in age-related neurodegenerative disorders such as Parkinson's and Alzheimer's disease. While mitophagy decreases throughout the lifespan of short-lived model organisms, it remains unknown whether such a decline occurs in the aging mammalian brain-a question of fundamental importance for understanding cell type- and region-specific susceptibility to neurodegeneration. Here, we define the longitudinal dynamics of basal mitophagy and macroautophagy across neuronal and non-neuronal cell types within the intact aging mouse brain in vivo. Quantitative profiling of reporter mouse cohorts from young to geriatric ages reveals cell- and tissue-specific alterations in mitophagy and macroautophagy between distinct subregions and cell populations, including dopaminergic neurons, cerebellar Purkinje cells, astrocytes, microglia and interneurons. We also find that healthy aging is hallmarked by the dynamic accumulation of differentially acidified lysosomes in several neural cell subsets. Our findings argue against any widespread age-related decline in mitophagic activity, instead demonstrating dynamic fluctuations in mitophagy across the aging trajectory, with strong implications for ongoing theragnostic development.

Keywords: Aging; Autophagy; Brain; Mitochondria; Mitophagy.

Figure 1. Mapping longitudinal changes in mitophagy and macroautophagy in the aging mammalian brain.

(A) Study overview. Mitophagy and macroautophagy decrease throughout the lifespan of short-lived experimental systems such as nematodes and yeast. Cell-specific alterations in mitophagy, and its relationship to macroautophagy in the aging mammalian brain are unknown. We examined the longitudinal progression of mammalian mitophagy and macroautophagy in natural aging using cohorts of reporter mice and quantified the spatiotemporal regulation of selective and nonselective turnover pathways in vivo using high-resolution confocal microscopy. (B) Mitophagy assay principle. The well characterized mito-QC mouse model (mCherry-GFP-FIS1^mt101-152) reports the abundance of mitolysosomes in mammalian cells and tissues. Functional mitochondria fluoresce in yellow (red and green) due to the endogenous mCherry-GFP tandem tag on their outer mitochondrial membrane. A pH-dependent fluorescence shift is observed when mitophagy occurs, as GFP fluorescence becomes quenched within the acidic endolysosomal network. This approach enables mitophagy to be quantified as mCherry-only puncta within cells and tissues. (C) Confocal tile scan of an axial brain section from the mito-QC mouse, showing widespread fluorescence. Scale bar 200 µm. (D) Macroautophagy assay principle. The auto-QC macroautophagy reporter mouse was made identical to the mito-QC mouse and maintained on the same genetic background. The auto-QC reporter works through the addition of MAPLC31B to the mCherry-GFP reporter. When nonselective macroautophagy occurs, autolysosomes are visualized through a pH-dependent fluorescence shift, enabling macroautophagy to be quantified as mCherry-only puncta within cells and tissues. (E) Confocal tile scan of an axial brain section from the auto-QC mouse, showing widespread fluorescence.

Figure 2. Temporal dynamics of regional mitophagy and macroautophagy in the CNS.

(A, B) PFC mitophagy. Representative confocal photomicrographs detailing instances of PFC mitophagy in young and geriatric mice. GFP and mCherry channels are shown for clarity, alongside quantitative analysis of PFC mitophagy levels across all ages. *P = 0.0188, ns = not significant; P > 0.9999. n = 47. (C, D) DG mitophagy. Representative confocal photomicrographs detailing instances of DG mitophagy in young and geriatric mice. GFP and mCherry channels are shown for clarity, alongside quantitative analysis of DG levels across all ages. ****P < 0.0001; ns = not significant; P = 0.0572. n = 47. (E, F) CB mitophagy. Representative confocal photomicrographs detailing instances of CB mitophagy in young and geriatric mice. GFP and mCherry channels are shown for clarity, alongside quantitative analysis of CB mitophagy levels across all ages. ****P < 0.0001. n = 47. Scale bar 20 µm. (G, H) PFC macroautophagy. Representative confocal photomicrographs detailing instances of PFC macroautophagy in young and geriatric auto-QC mice. GFP and mCherry channels are shown for clarity, alongside quantitative analysis of PFC macroautophagy levels across all ages. ****P < 0.0001. n = 27. (I, J) DG macroautophagy. Representative confocal photomicrographs detailing instances of DG macroautophagy in young and geriatric auto-QC mice. GFP and mCherry channels are shown for clarity, alongside quantitative analysis of DG macroautophagy levels across all ages. **P = 0.0011, ***P = 0.0006. n = 27. (K, L) CB macroautophagy. Representative confocal photomicrographs detailing instances of CB macroautophagy in young and geriatric auto-QC mice. GFP and mCherry channels are shown for clarity, alongside quantitative analysis of CB macroautophagy levels across all ages. ns P > 0.9999. n = 27. Scale bar 20 µm. Box plots extend from the 25th to the 75th percentiles, with a median line positioned inside the box. Whiskers denote the minimum and maximum values. Source data are available online for this figure.

Figure 3. Increased levels of neuronal mitophagy and differential lysosomal acidification define cerebellar aging in vivo.

(A) Schematic of a parasagittal brain section highlighting the cerebellum and Purkinje cells analyzed. (B) Mitophagy profiling in Purkinje neurons. Confocal photomicrographs from mito-QC brain sections showing representative instances of mitophagy throughout aging in cerebellar Purkinje neurons. Arrowheads indicate events of mitophagy. Scale bar = 20 µm. GCL granule cell layer, PCL Purkinje cell layer, ML molecular layer. (C) Macroautophagy profiling in Purkinje neurons. Confocal photomicrographs from auto-QC brain sections showing representative instances of nonselective macroautophagy throughout aging in cerebellar Purkinje neurons. Arrowheads indicate events of autophagy. Scale bar = 20 µm. (D) Quantitative analysis of Purkinje cells in mito-QC mice reveals a robust age-dependent increase in the number of acidified mitolysosomes in vivo, in addition to increases in mean mitolysosome size throughout time. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. n = 47. (E) Quantitative analysis of Purkinje cells in auto-QC mice demonstrates the numbers of acidified autolysosomes remain constant throughout aging and do not change in size. One-way ANOVA with Bonferroni post hoc. ns = not significant; P = > 0.9999. n = 27. (F) Identification of differentially acidified autolysosomes formed throughout aging within Purkinje cells in vivo. Immunohistochemical analysis of endolysosomal network using LAMP1 antibody identifies distinct lysosomal subpopulations. Shown is a 3D isosurface render generated using IMARIS, alongside confocal photomicrograph insets. Arrowheads indicate LAMP1-mCherry-positive (GFP-negative) colocalization, arrows indicate LAMP1-mCherry-GFP colocalization. Lysosomes are predominantly enriched at the pre-dendritic zones. Scale bar = 5 µm. (G) Quantitative analysis reveals an age-dependent increase in the number and mean size of differentially acidified lysosomes (LAMP1-mCherry-GFP-positive structures) over time. Although the overall number of LAMP1-positive endolysosomal structures remains constant throughout mammalian life, this compartment expands in size as a function of age. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999, ***P = 0.0002 and ****P = 0.0001. n = 27. (H) In vivo transmission electron microscopy (TEM) imaging of young and geriatric mito-QC mice. Scale bar = 1 µm. Quantitation of TEM images reveal an age-dependent increase in lysosomal (L) area, consistent with LAMP1 data in (G), with further TEM analysis demonstrating selective alterations in lysosomal area occurs at the somatodendritic zones of Purkinje neurons, while no change occurs at the axonal aspect. Mitochondria (Mit), Lipid droplet (LD). Student’s unpaired two-tailed t test. **P = 0.0019 and 0.0017; ns = not significant; P = 0.1225. n = 5. Box plots extend from the 25th to the 75th percentiles, with a median line positioned inside the box. Whiskers denote the minimum and maximum values. Source data are available online for this figure.

Figure 4. Spatiotemporal profiling of turnover pathways in aging mammalian hippocampal formation with emphasis on interneuronal subpopulations.

(A) Confocal tile scan of the hippocampal formation from the mito-QC mouse. Representative images from young and geriatric CA1 region. Scale bars = 20 µm. (B) Mitophagy in aging hippocampal formation. Quantitative analysis reveals a significant increase in CA1 mitophagy up until 15 months (****P value 0.0001), before a robust drop at geriatric age, returning to similar levels as in their young counterparts (ns P value 0.9999). The CA1 undergoes a significant increase in the number of differentially acidified lysosomes as a function of age (****P value 0.0001). n = 46. Quantitative analysis reveals no significant age-dependent changes in CA2 mitophagy, although a modest increase can be observed at 15-months (*P value 0.0272, ns = not significant; P value 0.0799). A robust increase in differentially acidified lysosomes can be observed in the CA2 as a function of age (****P value 0.0001). n = 46. Quantitative analysis reveals no significant changes in CA3 mitophagy throughout aging (ns P value 0.9999). A robust age-dependent increase in differentially acidified lysosomes can be observed in the CA3 (****P value 0.0001). n = 46. (C) Confocal tile scan of the hippocampal formation from the auto-QC mouse. Representative images from young and geriatric CA1 region. Scale bars = 20 µm. (D) Macroautophagy in the aging hippocampal formation. Quantitative analysis reveals a significant decline in CA1 macroautophagy as a function of age (****P value 0.0001, ***P value 0.0002). The CA1 also exhibits a significant age-dependent increase in the number of differentially acidified lysosomes (****P value 0.0001). n = 27. Quantitative analysis reveals that CA2 undergoes a robust decline in macroautophagy during aging (****P value 0.0001), with a simultaneous increase in differentially acidified lysosomes (****P value 0.0001). n = 27. Quantitative analysis reveals a significant age-dependent decline in CA3 macroautophagy (****P value 0.0001, **P value 0.0033). A significant increase in CA3 number of differentially acidified lysosomes can be seen as a function of age (****P value 0.0001). n = 27. (E) Representative maximum projections of mitophagy events in young and geriatric CA1 and dentate gyrus parvalbumin interneurons. Scale bar = 20 µm. (F) Mitophagy in CA1 parvalbumin interneurons. Quantitative analysis reveals no significant difference in young and geriatric CA1 interneuron mitophagy, although a small increase in levels can be seen through adulthood and mid-life (*P value 0.0335, ns = not significant; P value 0.9999). The CA1 interneurons undergo a robust increase in the number of differentially acidified lysosomes as a function of age (****P value 0.0001). n = 47. Mitophagy in dentate gyrus parvalbumin interneurons. Quantitative analysis reveals a robust increase in DG interneuron mitophagy up until 15-months before a substantial drop back to non-significant difference levels in geriatric mice compared to young (****P value 0.0001, ns = not significant; P value 0.9999). Dentate gyrus interneurons show a significant age-dependent increase in the number of differentially acidified lysosomes (***P value 0.0003). n = 46. (G) Representative maximum projections of macroautophagy events in young and geriatric CA1 and dentate gyrus parvalbumin interneurons. Scale bar = 20 µm. (H) Macroautophagy in CA1 parvalbumin interneurons. Quantitative analysis reveals a significant decline in CA1 interneuron macroautophagy thoughout adulthood and mid-life before at geriatric age returning to similar levels as their young counterparts (**P = 0.0018; ns = not significant; P = 0.3266). CA1 interneurons exhibit a robust increase in the number of differentially acidified lysosomes in geriatric mice compared to young (****P < 0.0001). n = 27. Macroautophagy in dentate gyrus parvalbumin interneurons. Quantitative analysis reveals a modest decline in dentate gyrus interneuron macroautophagy in geriatric mice compared to young (*P = 0.0252). Dentate gyrus interneurons exhibit a robust increase in the number of differentially acidified lysosomes in geriatric mice compared to young. (****P < 0.0001). n = 27. Box plots extend from the 25th to the 75th percentiles, with a median line positioned inside the box. Whiskers denote the minimum and maximum values. Source data are available online for this figure.

Figure 5. Temporal profiling of turnover pathways in aging mammalian dopaminergic circuits in vivo.

(A) Simplified schematic of A9 and A10 DA circuits. Both A9 and A10 neurons reside in the ventral mesencephalon (VM) and have distinct targets within the brain. A9 DA neurons project to the dorsolateral striatum, forming the nigrostriatal pathway that modulates voluntary locomotion, whereas A10 DA neurons project to the ventromedial striatum and forebrain to control reward, motivation, and aversion. (B, C) Mitophagy in aging A9 DA neurons. 3D Isosurface renders of young and geriatric mito-QC A9 dopaminergic neurons (scale bar = 5 µm) and representative confocal images of mitophagy events throughout aging, scale bar = 20 µm. Quantitative analysis reveals increased mitophagy levels as a function of age. One-way ANOVA with Bonferroni post hoc. **P = 0.0014. n = 43. (D, E) Macroautophagy in aging A9 DA neurons. Representative maximum projections of autophagy events throughout aging in A9 dopaminergic neurons (scale bar = 20 µm). Quantitative analysis of autophagy in A9 dopaminergic neurons shows no age-dependent alterations in autophagy levels. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 19. (F, G) Mitophagy in aging A10 DA neurons. Representative maximum projections of mitophagy events throughout aging in A10 dopaminergic neurons (scale bars = 20 µm). Quantitative analysis of mitophagy in A10 dopaminergic neurons reveals no change between young and geriatric mice. Some fluctuations in mitophagy levels were detected with a modest increase at 15 months. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999, *P = 0.024. n = 44. (H, I) Macroautophagy in aging A10 DA neurons. Isosurface renders of young and geriatric auto-QC A10 dopaminergic neurons (scale bar = 5 µm). Representative images of autophagy events throughout aging in A10 dopaminergic neurons (scale bar = 20 µm). Quantitative analysis of autophagy in A10 dopaminergic neurons reveals decreased levels of autophagy as a function of age. One-way ANOVA with Bonferroni post hoc. **P = 0.0076. n = 26. Box plots extend from the 25th to the 75th percentiles, with a median line positioned inside the box. Whiskers denote the minimum and maximum values. Source data are available online for this figure.

Figure 6. Temporal analysis of mitophagy and macroautophagy pathways in two anatomically and functionally distinct astroglial niches.

(A, B) Cerebellar astrocyte mitophagy. Representative images of mitophagy events throughout aging in cerebellar astrocytes in mito-QC mice (Overview, scale bar = 20 µm; close-up scale bar = 10 µm). Insets display mitolysosomes. Quantitative analysis reveals a robust increase in cerebellar astrocyte mitophagy levels as a function of age. Arrowheads indicate events of mitophagy. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001, ***P < 0.0002, **P = 0.0078. n = 46. (C, D) Cerebellar astrocyte macroautophagy. Representative images of autophagy events throughout aging in cerebellar astrocytes in auto-QC mice (Overview, scale bar = 20 µm; close-up scale bar = 10 µm). Quantitative analysis shows no significant changes in cerebellar astrocyte autophagy levels during aging. Arrowheads indicate events of autophagy. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (E, F) Dentate gyrus astrocyte mitophagy. Representative images of mitophagy events throughout aging in DG astrocytes in mito-QC mice (Overview, scale bar = 20 µm; close-up scale bar = 10 µm). Insets display mitolysosomes. Quantitative analysis reveals a modest increase in DG astrocyte mitophagy levels in geriatric mice compared to young. Arrowheads indicate events of mitophagy. One-way ANOVA with Bonferroni post hoc. *P = 0.0162. n = 47. (G, H) Dentate gyrus astrocyte macroautophagy. Representative images of autophagy events throughout aging in DG astrocytes in auto-QC mice (Overview, scale bar = 20 µm; close-up scale bar = 10 µm). Quantitative analysis shows no significant age-dependent changes in DG astrocyte autophagy levels during aging. Arrowheads indicate events of autophagy. One-way ANOVA with Bonferroni post hoc. ns = not significant; P = 0.1614. n = 27. Box plots extend from the 25th to the 75th percentiles, with a median line positioned inside the box. Whiskers denote the minimum and maximum values. Source data are available online for this figure.

Figure 7. Region-specific analysis of microglial mitophagy and macroautophagy pathways throughout healthy brain aging.

(A, B) Cerebellar microglial mitophagy. Representative images of mitophagy events throughout aging in cerebellar microglia in mito-QC mice (Overview, scale bar = 20 µm; close-up scale bar = 10 µm). Insets display mitolysosomes. Quantitative analysis reveals a robust increase in cerebellar microglia mitophagy levels as a function of age. Arrowheads indicate events of mitophagy. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. n = 46. (C, D) Cerebellar microglial macroautophagy. Representative images of autophagy events throughout aging in cerebellar microglia in auto-QC mice (Overview, scale bar = 20 µm; close-up scale bar = 10 µm). Quantitative analysis shows no significant changes in cerebellar microglia autophagy levels during aging, although a modest fluctuation in autophagy was detected at 16 months. Arrowheads indicate events of autophagy. One-way ANOVA with Bonferroni post hoc. ns = not significant; P = 0.6325, *P = 0.0131. n = 27. (E, F) Microglial mitophagy in the dentate gyrus. Representative images of mitophagy events throughout aging in dentate gyrus microglia in mito-QC mice (Overview, scale bar = 20 µm; close-up scale bar = 10 µm). Quantitative analysis reveals an increase in dentate gyrus microglia mitophagy levels throughout aging. Arrowheads indicate events of mitophagy. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. n = 47. (G, H) Microglial macroautophagy in the dentate gyrus. Representative images of autophagy events throughout aging in dentate gyrus microglia in auto-QC mice (Overview, scale bar = 20 µm; close-up scale bar = 10 µm). Quantitative analysis shows no significant age-dependent changes in dentate gyrus microglia autophagy levels during aging. Arrowheads indicate events of autophagy. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. Box plots extend from the 25th to the 75th percentiles, with a median line positioned inside the box. Whiskers denote the minimum and maximum values. Source data are available online for this figure.

Figure 8. Aging modifies CSF-producing choroid plexus turnover.

(A) Schematic of a parasagittal brain section highlighting the ChP within ventricle IV. (B, C) Mitophagy in the aging ChP. Representative images of mitophagy events throughout aging in the mito-QC choroid plexus. Quantitative analysis reveals increased levels of mitophagy in geriatric mice compared to young. Arrowheads indicate events of mitophagy. One-way ANOVA with Bonferroni post hoc. Scale bars = 20 µm. **P = 0.0287, *P = 0.045. n = 30. (D, E) Macroautophagy in the aging ChP. Representative images of autophagy events throughout aging in the auto-QC choroid plexus. Quantitative analysis reveals no age-dependent changes in autophagy through time. Arrowheads indicate events of autophagy. Scale bars = 20 µm. ns = not significant; P > 0.9999. n = 22. Box plots extend from the 25th to the 75th percentiles, with a median line positioned inside the box. Whiskers denote the minimum and maximum values. Source data are available online for this figure.

Figure 9. Spatiotemporal dynamics of mitophagy and macroautophagy in the brain throughout healthy aging.

Mitophagy and macroautophagy are dynamic processes that exhibit cell and region-specific changes throughout mammalian life. While mitophagy increases throughout lifespan in some cell types e.g., Purkinje neurons, decreases are observed in others, for instance in the dentate gyrus. Macroautophagy levels appear more stable across the brain than mitophagy levels. These spatiotemporal changes in autophagy pathways provide a valuable preclinical map for future phenotyping efforts and for developing targeted interventions that address impairments in cellular turnover, which can lead to neurological dysfunction and disease. Graphs are visual summaries and conceptual representations of relative trends per region and subset analyzed, with phenotypic summaries presented in the manuscript.

Figure EV1. Additional validation of optical reporter systems.

(A) Representative images of mito-QC human ARPE19 cells and Primary MEFs. (B) Representative images of auto-QC human ARPE19 cells and Primary MEFs. (C) Mitophagy in human mito-QC ARPE19 cells. One-way ANOVA with Tukey’s post hoc. ****P < 0.0001, ***P = 0.0001. n = 4. (D) Mitophagy in mito-QC primary MEFS. One-way ANOVA with Tukey’s post hoc. ****P < 0.0001. n = 4. (E) Mitophagy in adult mito-QC primary fibroblasts. One-way ANOVA with Tukey’s post hoc. ****P < 0.0001, ***P = 0.0005. n = 4. (F) Autophagy in auto-QC primary MEFS. One-way ANOVA with Tukey’s post hoc. ****P < 0.0001. n = 4. (G) Autophagy in human auto-QC ARPE19 cells. One-way ANOVA with Tukey’s post hoc. ****P < 0.0001, **P = 0.0022. n = 4. Source data are available online for this figure.

Figure EV2. Temporal dynamics of regional mitophagy and macroautophagy in the CNS—all channels (contains redisplay from Fig. 2).

(A, B) PFC mitophagy (redisplay of Fig. 2A with GFP channel). Representative confocal photomicrographs detailing instances of PFC mitophagy in young and geriatric mice. GFP and mCherry channels are shown for clarity, alongside quantitative analysis of % PFC mitophagy levels relative to mitochondrial area across all ages. Mean of 3-month group standardized to 100%. *P = 0.0489; ns = not significant; P > 0.9999. n = 47. (C, D) DG mitophagy (redisplay of Fig. 2C with GFP channel). Representative confocal photomicrographs detailing instances of DG mitophagy in young and geriatric mice. GFP and mCherry channels are shown for clarity, alongside quantitative analysis of % DG levels relative to mitochondrial area across all ages. Mean of 3-month group standardized to 100%. ****P < 0.0001, *P = 0.0338; ns = not significant; P = 0.1759. n = 47. (E, F) CB mitophagy (redisplay of Fig. 2E with GFP channel). Representative confocal photomicrographs detailing instances of CB mitophagy in young and geriatric mice. GFP and mCherry channels are shown for clarity, alongside quantitative analysis of % CB mitophagy levels relative to mitochondrial area across all ages. Mean of 3-month group standardized to 100%. ****P < 0.0001, **P < 0.0014. n = 47. Scale bar 20 µm. (G, H) PFC macroautophagy (redisplay of Fig. 2G with GFP channel). Representative confocal photomicrographs detailing instances of PFC macroautophagy in young and geriatric auto-QC mice. GFP and mCherry channels are shown for clarity, alongside quantitative analysis of % PFC macroautophagy levels relative to LC3 area across all ages. Mean of 3-month group standardized to 100%. ****P < 0.0001, ***P = 0.0003 **P = 0.0040. n = 27. (I, J) DG macroautophagy (redisplay of Fig. 2I with GFP channel). Representative confocal photomicrographs detailing instances of DG macroautophagy in young and geriatric auto-QC mice. GFP and mCherry channels are shown for clarity, alongside quantitative analysis of % DG macroautophagy levels relative to LC3 area across all ages. Mean of 3-month group standardized to 100%. ****P < 0.0001, **P = 0.0012, *P = 0.0261. n = 27. (K, L) CB macroautophagy (redisplay of Fig. 2K with GFP channel). Representative confocal photomicrographs detailing instances of PFC macroautophagy in young and geriatric auto-QC mice. GFP and mCherry channels are shown for clarity, alongside quantitative analysis of % CB macroautophagy levels relative to LC3 area across all ages. Mean of 3-month group standardized to 100%. ns = not significant; P > 0.9999. n = 27. Scale bar 20 µm. Box plots extend from the 25th to the 75th percentiles, with a median line positioned inside the box. Whiskers denote the minimum and maximum values. Source data are available online for this figure.

Figure EV3. Profiling of mitolysosome and autolysosome perimeter across the aging brain. Contains analysis from all cell-types and regions for completeness.

(A) Cerebellar mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. n = 47. (B) Cerebellar mean autolysosome perimeter. Quantitative analysis reveals no significant changes in autolysosome perimeter between geriatric mice compared to young. A modest decline in autolysosome perimeter is observed at 16 months. One-way ANOVA with Bonferroni post hoc. ns = not significant; P value > 0.9999, *P = 0.0297. n = 27. (C) Dentate gyrus mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. n = 47. (D) Dentate gyrus mean autolysosome perimeter. Quantitative analysis reveals a significant increase in autolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ***P = 0.0007. n = 27. (E) PFC mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. n = 47. (F) PFC mean autolysosome perimeter. Quantitative analysis reveals no significant changes in autolysosome perimeter between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (G) Purkinje cell mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. n = 47. (H) Purkinje cell mean autolysosome perimeter. Quantitative analysis reveals no significant changes in autolysosome perimeter between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P = 0.0657. n = 27. (I) Granular cell layer mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. n = 47. (J) Granular cell layer mean autolysosome perimeter. Quantitative analysis reveals no alterations in autolysosome perimeter between geriatric mice and young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (K) CA1 mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P value < 0.0001; *P = 0.0290. n = 47. (L) CA1 mean autolysosome perimeter. Quantitative analysis reveals no alterations in autolysosome perimeter between geriatric mice and young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P = 0.3350. n = 27. (M) CA2 mean mitolysosome perimeter. Quantitative analysis reveals fluctuations in mitolysosome perimeter in throughout lifespan. One-way ANOVA with Bonferroni post hoc. **P = 0.0094, *P = 0.0467; ns = not significant; P = 0.0875. n = 47. (N) CA2 mean autolysosome perimeter. Quantitative analysis reveals no alterations in autolysosome perimeter between geriatric mice and young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P value = 0.2924. n = 27. (O) CA3 mean mitolysosome perimeter. Quantitative analysis reveals fluctuations in mitolysosome perimeter in throughout lifespan. One-way ANOVA with Bonferroni post hoc. **P = 0.0034, *P = 0.0122; ns = not significant; P = 0.2975. n = 47. (P) CA3 mean autolysosome perimeter. Quantitative analysis reveals no alterations in autolysosome perimeter between geriatric mice and young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P = 0.8866. n = 27. (Q) CA1 Parvalbumin interneuron mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001, **P = 0.0015. n = 47. (R) CA1 Parvalbumin interneuron mean autolysosome perimeter. Quantitative analysis reveals no alterations in autolysosome perimeter between geriatric mice and young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P value > 0.9999. n = 27. (S) Dentate gyrus Parvalbumin interneuron mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001, **P = 0.0014. n = 46. (T) Dentate gyrus Parvalbumin interneuron mean autolysosome perimeter. Quantitative analysis reveals no alterations in autolysosome perimeter between geriatric mice and young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (U) A9 DA neuron mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. n = 43. (V) A9 DA neuron mean autolysosome perimeter. Quantitative analysis reveals no significant changes in autolysosome perimeter between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 19. (W) A10 DA neuron mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001; ***P = 0.0001. n = 44. (X) A10 DA neuron mean autolysosome perimeter. Quantitative analysis reveals no significant changes in autolysosome perimeter between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P = 0.1899. n = 26. (Y) Cerebellar astrocyte mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. ns = not significant; P = 0.9203. n = 46. (Z) Cerebellar astrocyte mean autolysosome perimeter. Quantitative analysis reveals no significant changes in autolysosome perimeter between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (AA) Dentate gyrus astrocyte mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. *P = 0.0165. n = 47. (AB) Dentate gyrus astrocyte mean autolysosome perimeter. Quantitative analysis reveals a modest increase in perimeter at midlife compared to young and returning to no significant change in geriatric mice. One-way ANOVA with Bonferroni post hoc. *P = 0.0240. ns = not significant; P = 0.1955. n = 27. (AC) Cerebellar microglia mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. n = 46. (AD) Cerebellar microglia mean autolysosome perimeter. Quantitative analysis reveals no significant changes in autolysosome perimeter between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (AE) Dentate gyrus microglia mean mitolysosome perimeter. Quantitative analysis reveals increased mitolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001 ***P = 0.0006. n = 47. (AF) Dentate gyrus microglia mean autolysosome perimeter. Quantitative analysis reveals a modest increase in autolysosome perimeter in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. *P = 0.00153. n = 27. (AG) ChP mean mitolysosome perimeter. Quantitative analysis reveals a modest increase in mitolysosome perimeter in geriatric mice compared to young. Some fluctuation with increased mitolysosome perimeter is observed at 15 months. One-way ANOVA with Bonferroni post hoc. ***P = 0.0006, *P = 0.0118. n = 30. (AH) ChP mean autolysosome perimeter. Quantitative analysis reveals no significant changes in autolysosome perimeter between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 22. Box plots extend from the 25th to the 75th percentiles, with a median line positioned inside the box. Whiskers denote the minimum and maximum values. Source data are available online for this figure.

Figure EV4. Profiling of mitolysosome and autolysosome eccentricity across the aging brain. Contains analysis from all cell-types and regions for completeness.

(A) Cerebellar mean mitolysosome eccentricity. Quantitative analysis reveals no significant changes in mitolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9771. n = 47. (B) Cerebellar mean autolysosome eccentricity. Quantitative analysis reveals no significant changes in autolysosome eccentricity between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P = 0.5185. n = 27. (C) Dentate gyrus mean mitolysosome eccentricity. Quantitative analysis reveals a modest increased mitolysosome eccentricity in during mature adulthood and midlife stages, before returning to corresponding eccentricity values in geriatric animals as in young. One-way ANOVA with Bonferroni post hoc. *P < 0.0436. ns = not significant; P = 0.3543. n = 47. (D) Dentate gyrus mean autolysosome eccentricity. Quantitative analysis reveals no significant changes in autolysosome eccentricity between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P = 0.9231. n = 27. (E) PFC mean mitolysosome eccentricity. Quantitative analysis reveals increased mitolysosome eccentricity in geriatric mice compared to young, indicating an elongated morphology as a function of age. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. n = 47. (F) PFC mean autolysosome eccentricity. Quantitative analysis reveals a modest increase in autolysosome eccentricity between geriatric mice compared to young animals, indicating mitolysosome elongation. Peak eccentricity values are observed at 16 months. One-way ANOVA with Bonferroni post hoc. *P = 0.0146, **P = 0.0035. n = 27. (G) Purkinje cell mean mitolysosome eccentricity. Quantitative analysis reveals increased mitolysosome eccentricity in geriatric mice compared to young. PC mitolysosome undergoes modest fluctuations in eccentricity throughout aging One-way ANOVA with Bonferroni post hoc. *P = 0.0170, **P = 0.0047. n = 47. (H) Purkinje cell mean autolysosome eccentricity. Quantitative analysis reveals no significant changes in autolysosome eccentricity between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (I) Granular cell layer mean mitolysosome eccentricity. Quantitative analysis reveals no significant changes in mitolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 47. (J) Granular cell layer mean autolysosome eccentricity. Quantitative analysis reveals no significant changes in autolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (K) CA1 mean mitolysosome eccentricity. Quantitative analysis reveals increased mitolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. **P = 0.0032. n = 47. (L) CA1 mean autolysosome eccentricity. Quantitative analysis reveals a significant decline in geriatric mice compared to young, indicating a more elongated shape with aging. One-way ANOVA with Bonferroni post hoc. ***P = 0.0009. n = 27. (M) CA2 mean mitolysosome eccentricity. Quantitative analysis reveals increased mitolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P value < 0.0001. *P = 0.0302. n = 47. (N) CA2 mean autolysosome eccentricity. Quantitative analysis reveals a significant decline in geriatric mice compared to young, indicating a more elongated shape with aging. One-way ANOVA with Bonferroni post hoc. **P = 0.0032. *P = 0.0237. n = 27. (O) CA3 mean mitolysosome eccentricity. Quantitative analysis reveals increased mitolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ****P < 0.0001. ***P = 0.0003. n = 47. (P) CA3 mean autolysosome eccentricity. Quantitative analysis reveals no significant change in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P = 0.8531. n = 27. (Q) CA1 Parvalbumin interneuron mean mitolysosome eccentricity. Quantitative analysis reveals increased mitolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ***P = 0.0004. *P = 0.0350. n = 47. (R) CA1 Parvalbumin interneuron mean autolysosome eccentricity. Quantitative analysis reveals no significant changes in autolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (S) Dentate gyrus Parvalbumin interneuron mean mitolysosome eccentricity. Quantitative analysis reveals no significant change in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 46. (T) Dentate gyrus Parvalbumin interneuron mean autolysosome eccentricity. Quantitative analysis reveals no significant change in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (U) A9 DA neuron mean mitolysosome eccentricity. Quantitative analysis reveals increased mitolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.0689. n = 43. (V) A9 DA neuron mean autolysosome eccentricity. Quantitative analysis reveals no significant changes in autolysosome eccentricity between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 19. (W) A10 DA neuron mean mitolysosome eccentricity. Quantitative analysis reveals increased mitolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.0784. n = 44. (X) A10 DA neuron mean autolysosome eccentricity. Quantitative analysis reveals no significant changes in autolysosome eccentricity between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 26. (Y) Cerebellar astrocyte mean mitolysosome eccentricity. Quantitative analysis reveals no significant change in mitolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 46. (Z) Cerebellar astrocyte mean autolysosome eccentricity. Quantitative analysis reveals no significant changes in autolysosome eccentricity between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (AA) Dentate gyrus astrocyte mean mitolysosome eccentricity. Quantitative analysis reveals increased mitolysosome eccentricity in geriatric mice compared to young animals. One-way ANOVA with Bonferroni post hoc. ***P = 0.0005. **P = 0.0041. n = 47. (AB) Dentate gyrus astrocyte mean autolysosome eccentricity. Quantitative analysis reveals no significant changes between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P value > 0.9999. n = 27. (AC) Cerebellar microglia mean mitolysosome eccentricity. Quantitative analysis reveals increased mitolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.7837. n = 46. (AD) Cerebellar microglia mean autolysosome eccentricity. Quantitative analysis reveals no significant changes in autolysosome eccentricity between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (AE) Dentate gyrus microglia mean mitolysosome eccentricity. Quantitative analysis reveals no significant change in mitolysosome eccentricity in geriatric mice compared to young animals, although a modest increase can be observed at 15 months One-way ANOVA with Bonferroni post hoc. *P = 0.0328, ns = not significant; P > 0.5370. n = 47. (AF) Dentate gyrus microglia mean autolysosome eccentricity. Quantitative analysis reveals no significant changes in autolysosome eccentricity between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (AG) ChP mean mitolysosome eccentricity. Quantitative analysis reveals increased mitolysosome eccentricity in geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant; P = 0.0922. n = 30. (AH) ChP mean autolysosome eccentricity. Quantitative analysis reveals no significant changes in autolysosome eccentricity between geriatric mice compared to young. One-way ANOVA with Bonferroni post hoc. ns = not significant P > 0.9999. n = 22. Box plots extend from the 25th to the 75th percentiles, with a median line positioned inside the box. Whiskers denote the minimum and maximum values. Source data are available online for this figure.

Figure EV5. Mitophagy and macroautophagy in cerebellar granule neurons in vivo.

(A) Purkinje cell acidified autolysosomes. The number of LAMP1-positive structures that are double-positive for mCherry-only remains unchanged between young and geriatric mice. Some fluctuations of increased acidified autolysosomes can be observed at 16 months compared to the geriatric. One-way ANOVA with Bonferroni post hoc. **P < 0.0097, ns = not significant; P > 0.0829. n = 27. (B) Purkinje cell mean acidified autolysosome size. mCherry-LAMP1 double-positive structure size remain constant throughout aging. One-way ANOVA with Bonferroni post hoc. ns = not significant; P > 0.9999. n = 27. (C–E) Mitophagy in the aging GCL. Representative images of mitophagy events throughout aging in the mito-QC granular cell layer. Quantitative analysis reveals increased levels of mitophagy in geriatric mice compared to young, as do mean mitolysosome size. One-way ANOVA with Bonferroni post hoc. Scale bars = 20 µm. ****P < 0.0001. n = 47. (F–H) Autophagy in the aging GCL. Representative images of autophagy events throughout aging in the auto-QC granular cell layer. Quantitative analysis reveals unaltered levels of autophagy in geriatric mice compared to young, nor do autolysosome size change. One-way ANOVA with Bonferroni post hoc. Scale bars = 20 µm. ns = not significant; P > 0.9999. n = 27. Box plots extend from the 25th to the 75th percentiles, with a median line positioned inside the box. Whiskers denote the minimum and maximum values. Source data are available online for this figure.