Dasatinib and Quercetin Reverse Retinal Aging in Dry AMD Mouse Model

Dry age-related macular degeneration (AMD) is the leading cause of irreversible central vision loss in older adults, yet no disease-modifying treatments exist. Its hallmarks — lipofuscin accumulation in the retinal pigment epithelium (RPE), RPE degeneration, and photoreceptor atrophy — have been linked to cellular senescence, but whether clearing senescent cells could slow disease progression had not been directly tested. This study is the first to evaluate long-term senolytic therapy with dasatinib and quercetin (D+Q) in a preclinical dry AMD model, providing a mechanistic framework connecting RPE senescence, lysosomal dysfunction, and lipofuscin toxicity.

The study used Abca4−/−Rdh8−/− double-knockout mice on a C57BL/6J background — an established dry AMD model in which disrupted visual cycle proteins cause progressive lipofuscin buildup and retinal degeneration. Starting at 4 months of age, mice received oral gavage of 5 mg/kg dasatinib plus 50 mg/kg quercetin on three consecutive days every two weeks for 4 months (vehicle controls received 30% PEG400). All animals were assessed at 8 months. Human relevance was established by analyzing the GEO dataset GSE29801, comprising bulk RNA-seq from RPE of geographic atrophy patients versus age-matched controls, using GSEA and differential expression analysis.

In the human transcriptomic analysis, RPE from geographic atrophy patients showed significant upregulation of SASP factors (interleukins, chemokines, MMPs) and senescence markers CDKN1A and CDKN2A. GSEA confirmed enrichment of senescence gene sets: the SAL_SEN_MAYO set achieved a normalized enrichment score (NES) of 2.07 (FDR q<0.001) and DEMAGALHAES_AGING_UP reached NES 1.46 (FDR q=0.04). Lysosomal acidification gene sets and oxidative stress response pathways were also significantly downregulated in geographic atrophy RPE, establishing a human disease context for the mouse intervention.

In the mouse model, 8-month-old untreated Abca4−/−Rdh8−/− mice exhibited robust RPE senescence confirmed by elevated SA-β-galactosidase activity, increased p16 and p21 protein expression, and upregulated SASP cytokines. Lysosomal pH was significantly elevated (alkalinized) compared to wild-type controls, and lipofuscin autofluorescence in RPE flat mounts was markedly increased. Oxidative stress markers — including reduced total antioxidant capacity, decreased GSH/GSSG ratio, and elevated protein carbonyls — were all significantly worsened. Structural analysis by H&E staining showed thinning of the photoreceptor outer nuclear layer (ONL), and transmission electron microscopy revealed RPE ultrastructural damage. Electroretinography (ERG) demonstrated reduced scotopic a-wave, b-wave, and c-wave amplitudes, as well as diminished photopic responses, indicating functional vision loss.

Long-term D+Q treatment significantly reversed these pathological changes. Treated mice showed reduced SA-β-Gal staining, lower p16/p21 expression, and suppressed SASP cytokine levels in RPE. Lysosomal pH was restored toward normal, lipofuscin autofluorescence was substantially reduced, and ZO-1 tight junction integrity in the RPE monolayer was better preserved. Antioxidant capacity improved, GSH/GSSG ratios recovered, and protein carbonylation decreased. Structurally, ONL thickness was better maintained, and ERG amplitudes were significantly higher in D+Q-treated mice compared to vehicle controls. These findings collectively demonstrate that senolytic clearance of RPE senescent cells interrupts a pathological cascade linking senescence → lysosomal dysfunction → lipofuscin accumulation → oxidative stress → retinal degeneration.

The translational implications are meaningful. Dasatinib is already FDA-approved (for leukemia) and quercetin is widely available as a supplement, and the intermittent dosing regimen used here (3 days on, 11 days off, every 2 weeks) mirrors protocols already being tested in human clinical trials for other age-related conditions. This study provides the first direct preclinical evidence that this combination can slow dry AMD progression, opening a potential therapeutic avenue for a disease with no current disease-modifying options. Limitations include the exclusively mouse-based intervention data and the absence of dose-optimization or safety profiling specific to ocular tissues.