The retina was used as an experimentally accessible tissue because it combines high metabolic demand with tightly regulated interactions among neurons, glia and resident immune cells. Electroretinography, or ERG, provides a quantitative readout of electrical responses from several retinal components. The investigators combined this functional measurement with biochemical assays and immunohistochemistry to ask whether mitochondrial genetic background could modify retinal function, mitochondrial composition and the abundance and localization of STING, a protein involved in innate immune signaling.

The study included mice approximately 2, 10 and 20 months old, while ERG recordings and retinal ATP measurements were restricted to younger animals because of study-design constraints. To isolate mitochondrial genetic effects from nuclear genetic effects, the researchers used mitochondrial–nuclear exchange, or MNX, mice. All animals had a C57BL/6J nuclear genome, but one group carried C57BL/6J mitochondrial DNA and the other carried C3H/HeN mitochondrial DNA.

The two mitochondrial genomes differ by two missense variants: one in a subunit of respiratory-chain complex I and one in a subunit of complex IV. Approximately 113 male and female mice were used across the experiments. Most molecular and imaging groups contained only three or four animals, making the work exploratory and limiting statistical power, particularly for interaction tests and sex-specific analyses.

Western blotting revealed that aging changed the abundance of proteins associated with oxidative phosphorylation, but the pattern varied by mitochondrial background. In mice carrying C3H mitochondrial DNA, the measured complex I subunit NDUFB8 increased between 2 and 10 months and remained elevated at 20 months, while it remained stable in the C57 mitochondrial-DNA group. Complex II showed a similar distinction: SDHB abundance increased progressively across the three ages in the C3H group but did not change in the C57 group.

Complex III declined markedly with age in both backgrounds. Complex IV was lower at 2 months with C3H mitochondrial DNA, increased with age and surpassed the corresponding C57 level by 20 months. The ATP synthase alpha subunit, representing complex V, was also lower at baseline in the C3H group and decreased significantly in both backgrounds by 10 months.

The respiratory-chain pattern is consistent with age-related remodeling rather than a uniform decline in every mitochondrial component. Complex III is a major site of reactive oxygen species production through the Q cycle, making its age-related reduction potentially relevant to oxidative stress. Complex IV rose while complex V fell, a combination that could reflect compensatory maintenance of electron transport when ATP production is constrained.

These remain interpretations rather than demonstrated mechanisms: the investigators did not directly quantify reactive oxygen species, transcriptional feedback, glycolytic reliance or mitochondrial mass. The abundance measurements were normalized to beta-actin, a cytoskeletal protein, and the authors note that accounting directly for changes in mitochondrial abundance would sharpen interpretation of the observed remodeling. ERG measurements at 3 weeks and 2 months showed that retinal function was not uniformly impaired in either mitochondrial background.

Photoreceptor and retinal ganglion-cell responses were broadly comparable, but the b-wave, which reflects bipolar-cell activity modulated by Müller glia, showed an interaction between mitochondrial background and stimulus intensity. C57 mitochondrial-DNA mice had lower b-wave amplitudes than C3H mitochondrial-DNA mice at 3 weeks. The c-wave, which the study used to assess retinal pigment epithelium activity, was higher at 2 months than at 3 weeks regardless of mitochondrial background in the reported comparison, with a significant age difference identified in the C3H group.

Total retinal ATP measured at 8 weeks did not differ significantly between backgrounds, indicating that the respiratory-protein differences did not produce a detectable difference in this bulk ATP measure. A quantitative PCR assay compared amplification of long and short mitochondrial-DNA fragments to estimate mitochondrial-DNA lesions. Lesion burden increased significantly with age in the C57 mitochondrial-DNA group, whereas the C3H group did not show a significant age-related increase.

At 20 months, the C3H group had a lower lesion burden than the C57 group. Because mitochondrial DNA can function as a damage-associated signal when released into the cytosol, this divergence could be relevant to cGAS–STING signaling. The experiment did not, however, directly measure cytosolic mitochondrial DNA, cGAS activity, interferon responses or downstream STING activation.

The proposed connection between accumulated lesions and inflammatory signaling therefore remains unproven in this study. Immunohistochemical analysis showed that microglial STING abundance increased strongly with age in untreated retinas: 20-month-old animals had more STING within microglia than the younger groups. The overall analysis of untreated tissue did not find a significant main effect of mitochondrial background or a significant age-by-background interaction.

A direct comparison nevertheless found a lower STING burden in 20-month-old C3H mitochondrial-DNA mice. Acute elevation of intraocular pressure produced a different pattern. In a mixed-effects analysis that accounted for contralateral eyes from the same animal, treatment, age and an age-by-genotype interaction significantly influenced microglial STING abundance.

Pressure elevation reduced STING abundance in 20-month-old C57 animals relative to their control eyes, while other treatment comparisons did not reach statistical significance. Age was also associated with a change in the intracellular distribution of STING. Colocalization between STING and LAMP1, a lysosomal marker, was lower in 20-month-old mice than in either younger group.

This age effect did not differ significantly by mitochondrial background and was not significantly changed by acute pressure elevation. The imaging result is compatible with altered STING trafficking or lysosomal processing, including possible changes in degradation, but the study did not directly test those mechanisms. It also quantified protein abundance and localization rather than functional pathway activity, so the reduced lysosomal association cannot be interpreted as proof that aged microglia had stronger or weaker STING signaling.

Measurements of interferon responses, additional pathway components, multiple recovery intervals and chronic pressure paradigms would be required to address those questions. Overall, the study identifies coordinated but not necessarily causal changes in retinal metabolism and microglial protein regulation across aging in two mitochondrial–nuclear exchange mouse lines. C3H mitochondrial DNA was associated with age-related increases in complexes I and II, less mitochondrial-DNA damage at 20 months and a different pattern of microglial STING abundance under control and pressure-stressed conditions.

Age increased microglial STING and reduced STING–LAMP1 colocalization across both backgrounds. Important limitations remain: ERG data were unavailable at 10 and 20 months, several cohorts were small, sex effects could not be assessed with adequate power in most experiments, and only one hour of recovery was examined after acute pressure elevation.

Subject of Research: Effects of mitochondrial genetic background on retinal aging, bioenergetics and microglial STING in mice

Article Title: Mitochondrial Genetic Background Shapes Baseline Retinal Function, Age‐Related Bioenergetic Remodeling, and Microglial STING Phenotypes

Article References: Somerville, M. M., Lin, S. S., Ballinger, S. W., & Gross, A. K. (2026). Mitochondrial Genetic Background Shapes Baseline Retinal Function, Age‐Related Bioenergetic Remodeling, and Microglial STING Phenotypes. Aging Cell, 25(10), Article e70771. https://doi.org/10.1111/acel.70771

Image Credits: AI Generated

DOI: 10.1111/acel.70771

Keywords: retinal aging, mitochondrial DNA, oxidative phosphorylation, microglia, STING signaling, mitochondrial damage, electroretinography, mouse model, intracellular pressure, innate immunity, Mitochondrial, Genetic