H3K9me3 inhibition reverses Alzheimer′s progression by restoring synaptic and immune proteostasis across the brain–retina axis

H3K9me3 inhibition reverses Alzheimer′s progression by restoring synaptic and immune proteostasis across the brain–retina axis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article H3K9me3 inhibition reverses Alzheimer′s progression by restoring synaptic and immune proteostasis across the brain–retina axis Maya Koronyo-Hamaoui, Dieu-Trang Fuchs, Jean-Philippe Vit, Altan Rentsendorj, and 17 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8913130/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Epigenetic dysregulation is increasingly linked to ageing and neurodegeneration, yet its contribution to retinal and brain pathology in Alzheimer's disease (AD) remains uncharacterized. We quantified the repressive histone mark H3K9me3 in post-mortem retinas and brains from donors spanning normal cognition, mild cognitive impairment, and AD dementia. In both tissues, H3K9me3 increased in early stages and further in AD dementia, strongly associating with cognitive status and neuropathological burden. To assess causality, we inhibited the H3K9 methyltransferase SUV39H1 in APPswe/PS1dE9 and APPswe/tauP301L/PS1tm1Mpm mouse models. SUV39H1 inhibition lowered H3K9me3, mitigated AD-like pathology, restored synaptic integrity, and improved cognitive and visual performance. Proteomics revealed that H3K9me3 derepression reestablished retinal and brain proteostasis and promoted neuroprotection through immunomodulatory pathways and BDNF/VGF–granin signalling. These findings identify H3K9me3 as shared epigenetic driver of AD-related dysfunction, highlight H3K9me3 reduction as therapeutic strategy, and position the retina as an accessible extension of the brain for epigenetic studies. Biological sciences/Neuroscience/Epigenetics in the nervous system/Epigenetics and behaviour Biological sciences/Neuroscience/Epigenetics in the nervous system/Epigenetics and plasticity Neurodegenerative disease heterochromatin histone modification NT1721 ocular manifestations granin family members Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Main Epigenetic processes are increasingly recognized as key contributors to ageing and neurodegenerative disorders, including Alzheimer’s disease (AD) 1-8 , the most common cause of senile dementia 9 . Canonical neuropathological hallmarks of AD, amyloid β-protein (Aβ) plaques and neurofibrillary tangles composed of hyperphosphorylated tau, together with neuroinflammation, vascular injury, and progressive synaptic and neuronal loss 10-12 , extend beyond the brain to the neurosensory retina 13-27 . Although retinal epigenome undergoes dynamic remodelling with age and disease 28-31 , epigenetic mechanisms underlying AD-associated retinal dysfunction remain largely uncharacterized. Because the retina is an anatomically accessible extension of the central nervous system (CNS), defining epigenetic signatures that track AD progression at the brain–retina interface could enable noninvasive biomarker development for longitudinal monitoring and inform precision epigenetic interventions. Histone methylation and acetylation are central regulators of gene expression across the lifespan 32,33 and essential for synaptic plasticity as well as learning and memory formation 34-37 . Among these modifications, repressive chromatin marks—di- and tri-methylation of histone H3 lysine 9 (H3K9me2/3)—are increased in AD patient brains and in AD mouse models 38-40 . Notably, elevated H3K9me3, catalysed by the histone methyltransferase SUV39H1, is linked to transcriptional silencing of genes critical for synaptic function 39 . Given the reversibility of epigenetic modifications, histone-modifying enzymes represent promising therapeutic targets for preserving neuronal function during ageing and neurodegeneration 41,42 . However, whether targeting H3K9me3 can mitigate AD-related dysfunction in the brain and retina, and thereby preserve cognitive and visual function, remains unknown. In this study, we profiled H3K9me3 and related proteomic pathways in postmortem brains and corresponding retinas of cognitively normal (CN) individuals and patients with mild cognitive impairment (MCI due to AD) or AD dementia. We found a strong association between retinal and brain H3K9me3 levels, AD pathology, and cognitive status— starting to elevate at the earliest clinical stage (MCI). To define the functional contribution of this repressive mark, we pharmacologically inhibited the H3K9 methyltransferase SUV39H1, using the synthetic epidithiodiketopiperazine compound ETP69 (NT1721) 43 , in two transgenic mouse models of AD (APPswe/PS1dE9 and APPswe/tauP301L/PS1 tm1Mpm ). By integrating brain–retina biochemical and proteomic profiling, quantitative histology, blood-based immunophenotyping, and behavioural analyses, pharmacological inhibition of SUV39H1 reduced H3K9me3 burden, attenuated AD-like neuropathology, restored synaptic integrity, and rescued both cognitive performance and visual function. Mechanistically, ETP69 conferred neuroprotection by re-establishing retina and brain proteostasis, accompanied by coordinated remodelling of pathways governing innate immune activation and trophic support, including BDNF–VGF–granin signalling. Results Elevated cerebral and retinal H3K9me3 associates with neuropathology and cognitive impairment in MCI/AD H3K9me3 levels in the brain and retina were measured in patients with a clinical and post-mortem neuropathological diagnosis of AD dementia (n=13, mean age 85.5 ± 2.7 years, 7 females/6 males) or MCI (due to AD; n=7, mean age 91.1 ± 1.9 years, 4 females/3 males), and compared to age- and sex-matched CN individuals (n=6, mean age 91.3 ± 2.8 years, 4 females/2 males; human donor information is summarized in Table 1 and Extended Data Table 1). In the dorsolateral prefrontal cortex (Brodmann area 9; A9), a region critical for executive function and affected in AD, immunohistochemical (IHC) quantification of H3K9me3 immunoreactive (IR) area normalised to DAPI nuclear density revealed robust increases in both disease groups, with significant 3.3-fold and 2.2-fold elevations in AD and MCI, respectively, relative to CN controls (AD, p < 0.0001; MCI, p = 0.0207; Fig. 1a,b; H3K9me3 raw area analysis in Extended Data Fig. 1a). No difference between male and female patients was detected (Extended Data Fig. 1b). Higher cortical H3K9me3 strongly correlated with cognitive deficits, with higher CDR) and lower MMSE scores (CDR: r = 0.68, p = 0.0004; MMSE: r = –0.82, p < 0.0001; Fig. 1c,d). In the cortical A9 region, H3K9me3 levels more strongly correlated with NFT severity ( r = 0.75, p = 0.0001; Fig. 1e) than Aβ-plaque burden ( r = 0.57, p = 0.0070; Fig. 1f). Cortical H3K9me3 further associated with AD neuropathological indices, disease stage, and cognitive performance across analyses stratified as CN+MCI, MCI+AD, or all groups combined (heatmap summary in Extended Data Fig. 1c). In the retina from the same cohort, quantitative histology of H3K9me3 per nuclei showed significant 2.2- and 1.6-fold increases in AD and MCI patients, respectively, compared to CN individuals (p < 0.0001 and p = 0.0087; Fig. 1g,h; H3K9me3 raw area analysis in Extended Data Fig. 1d), with no difference between male and female AD patients (Extended Data Fig. 1e). H3K9me3 immunoreactivity was detected across all retinal nuclear layers, with particularly strong signal in the retinal ganglion cell layer and inner nuclear layer, and enrichment in cells proximal to Aβ deposits, most evident in MCI and AD retinas (Fig. 1g). Elevated retinal H3K9me3 closely tracked retinal 4G8 + Aβ deposition (r = 0.85, p < 0.0001) and, albeit to a lesser extent, remained strongly associated with retinal T22 + tau oligomers (r = 0.65, p = 0.0051; Fig. 1i). Retinal H3K9me3 also correlated with retinal ionized calcium-binding adaptor molecule 1 (IBA1) + microgliosis and glial fibrillary acidic protein (GFAP) + macrogliosis ( r = 0.64, p = 0.0046 and r = 0.62, p = 0.0102, respectively; Extended Data Fig. 1f). Notably, retinal H3K9me3 tightly correlated with cortical H3K9me3 ( r = 0.70, p = 0.0004; Fig.1j), and associated with cognitive status (CDR: r = 0.65, p = 0.0008 and MMSE: r = –0.62, p = 0.0042; Fig. 1k). These data suggest a coordinated brain–retina increase in the repressive chromatin mark H3K9me3 that tracks AD neuropathology, disease stage, and cognitive impairment. Proteomics identifies SUV39H1–H3K9 hypertrimethylation in AD brain and retina To extend our histological findings and define perturbations in epigenetic pathways across the AD brain and retina, with a focus on H3K9 regulation, we performed a targeted reanalysis of mass spectrometry–based proteomic datasets from independent human cohorts 16 (temporal cortex: n=10 AD, n=8 CN; temporal hemiretina: n=6 AD, n=6 CN; Fig. 1l–o). Metascape Gene Ontology analysis of differentially expressed proteins (DEPs) in AD brains, revealed significant enrichment of pathways governing epigenetic regulation, chromatin organization, nucleosome assembly, and histone modification (Fig. 1l). Importantly, ingenuity pathway analysis (IPA) predicted SUV39H1 activation, by expression profiles of directly and indirectly connected DEP network in AD cortices (Fig. 1m), consistent with elevated cortical H3K9me3 observed in MCI and AD brains. Histone variants were broadly increased, including H1, which stabilizes linker DNA and promotes higher-order chromatin compaction, H3-3, which increases with ageing 44-46 , and H2AZ1, a transcriptional repressor of activity-dependent plasticity genes 47 (Extended Data Fig. 2a,b). Enzymes regulating histone arginine methylation were also altered: PRMT1, which methylates H4R3 and supports H3K9 acetylation, was reduced, whereas PRMT5, which methylates H3R8 and antagonizes H3K9 acetylation, was increased, collectively shifting chromatin toward reduced H3K9ac and enhanced H3K9 methylation (Extended Data Fig. 2a,b). Proteomic profiling of AD retinas similarly revealed dysregulation of pathways linked to epigenetic control, chromatin organization, heterochromatin formation, and histone H3 methyltransferase activity (Fig. 1n and Extended Data Fig. 2c). The expression pattern of histone modifying enzymes in AD retina predicted enhanced H3K9 hypermethylation (Fig. 1o). Lysine acetyltransferase 2B (KAT2B) and TATA-box binding protein associated factor 1 (TAF1), which acetylate H3K9, and the H3K9me3-specific lysine demethylases 4B and 4C (KDM4B/C), were downregulated, collectively promoting H3K9 trimethylation. Lysine acetyltransferase 7 (KAT7), which restricts SUV39H1-mediated heterochromatin spreading 48 , was also reduced, whereas the arginine demethylase and lysine hydroxylase Jumonji domain containing 6 (JMJD6), which demethylates H4R3 49 and can bias chromatin toward H3K9 methylation, was increased. Together with the histological findings, these proteomic signatures converge on elevated H3K9me3 in both AD brain and retina, supporting H3K9me3 as a cross-compartment epigenetic correlate of disease progression and providing a mechanistic rationale for targeting SUV39H1-heterochromatinization as a therapeutic strategy in AD. SUV39H1 inhibition rescues cognitive and visual function in aged AD models We next tested whether ETP69-mediated SUV39H1 inhibition improves cognitive and visual performance in 18-month-old APPswe/PS1dE9 (AD⁺) mice and age- and sex-matched WT littermates (Fig. 2a–h and Extended Data Figs. 3,4). Mice received intraperitoneal ETP69 (10 mg kg⁻¹) under distinct dosing regimens (n = 118), whereas controls received vehicle (DMSO), and all animals completed a behavioural battery spanning locomotion, hippocampus-dependent spatial learning and memory, and visual stimulus discrimination to capture both rapid and sustained effects from 1 to 14 days post-treatment (Fig. 2a,b and Extended Data Figs. 3a,4a). Short-term testing (days 1–3; open field, Y-maze, and colour- and contrast-mode visual X-maze) showed that ETP69 did not alter locomotor activity in either genotype (distance travelled, resting time, average speed, rearing, or arm entries; Fig. 2c and Extended Data Fig. 3b,c). Notably, ETP69 reversed working-memory deficits in aged AD⁺ mice, increasing spontaneous alternations across assays (Y-maze: p = 0.0093; colour X-maze: p < 0.0001; contrast X-maze: p = 0.0020; Fig. 2d; repeated-dose data in Extended Data Fig. 3d). A single dose also increased alternations in aged WT mice in the colour-mode X-maze (p = 0.0027; Fig. 2d). Repeated ETP69 dosing rescued colour-mode X-maze visual performance in AD⁺ mice, restoring blue-arm entries and blue↔white bidirectional transitions to WT-like levels (baseline deficits: p = 0.0059 and p = 0.0009; rescue: p = 0.0383 and p = 0.0035; Fig. 2e and Extended Data Fig. 3e,f). We further examined whether ETP69 elicited sustained improvements in hippocampus-dependent learning and memory from days 4–14 using the Barnes maze and contextual fear conditioning (combined data in Fig. 2f–h; regimen-specific data in Extended Data Fig. 4). In the Barnes maze, vehicle-treated AD⁺ mice made more errors than WT across all phases (acquisition: repeated measures (RM) ANOVA, F(1,30) = 27.40, p < 0.0001; retention: p < 0.0001; reversal: F(1,23) = 34.07, p < 0.0001), whereas ETP69 reduced errors during acquisition (F(1,27) = 10.23, p = 0.0035), retention (p < 0.0001), and reversal (F(1,20) = 16.19, p = 0.0007), restoring performance to near WT levels (Fig. 2f and Extended Data Fig. 4b). Movement-transition chord diagrams further supported this rescue, showing focused search near the escape box in ETP69-treated AD⁺ mice during retention (red ribbons) and reversal (blue ribbons), in contrast to the non-goal-oriented exploration of vehicle-treated AD⁺ mice (Fig. 2g). Associative learning and memory assessed by contextual fear conditioning showed increased freezing 24 h after conditioning in both genotypes following ETP69 treatment (WT: p = 0.0119 in the first minute; AD⁺: RM ANOVA, F(1,26) = 11.71, p = 0.0021; Fig. 2h and Extended Data Fig. 4c,d). Overall, ETP69 produced rapid and sustained improvements in spatial learning, long-term memory retention, reversal learning, colour vision, and associative memory in aged AD⁺ mice. Performance did not differ by sex across Y-maze, colour- and contrast-mode X-maze, and Barnes maze (Extended Data Fig. 4e); accordingly, males and females were pooled for subsequent molecular analyses. Cerebral H3K9me3 reduction curbs Aβ pathology and astrogliosis while preserving synaptic integrity Given the behavioural benefits of SUV39H1 inhibition, we next examined its impact on AD-associated neuropathology. Immunohistochemistry and immunoblotting of 18-month-old AD + mouse brains showed elevated H3K9me3 versus WT littermates, with a 1.3–1.5-fold increase in cortex (p < 0.0001; Fig. 2i), including layers II/III and VI (Extended Data Fig. 5a,b), and higher levels across the hippocampal formation (p = 0.0782 to p < 0.0001; Fig. 2j,k), spanning CA1–CA3 and the dentate gyrus (DG) (Extended Data Fig. 5c). A single ETP69 injection markedly lowered H3K9me3 in AD + mice as early as 4 days post-injection, reducing levels by 40% in cortex (p < 0.0001; Fig. 2i) and by 32% in hippocampus (p = 0.0454; Fig. 2j) to below WT control levels. These effects persisted 15 days post-injection, with 40% reduction in cortex (p= 0.0069) and 47% in hippocampus (CA: 54% and DG 42%, p< 0.0001; Fig. 2i,k and Extended Data Fig. 5c). In aged WT mice, ETP69 also significantly lowered H3K9me3 by 27% in cortex (p = 0.0004; Fig. 2i) and 36% in hippocampus at day 4 (p = 0.0286; Fig. 2j), and 36% in the CA region at day 15 (p= 0.0185; Extended Data Fig. 5c). Evaluation of core AD pathology and associated gliosis indicated attenuation of Aβ plaque burden and GFAP reactive astrogliosis following ETP69 treatment. In cortex, ETP69 caused marked reductions in GFAP + and 6E10-Aβ plaques at both 4 days (GFAP: 58%, p < 0.0001; 6E10: 42%, p = 0.0004) and 15 days post-injection (GFAP: 44%, p = 0.0133; 6E10: 66%, p = 0.0004; Fig. 2l). In hippocampus, while vehicle-treated AD + mice exhibited pronounced astrogliosis compared to WT mice (3.3-fold, p < 0.0001), ETP69 significantly attenuated GFAP levels 4 days after (29%, p = 0.0223; Fig. 2m) and remained reduced 15 days post-treatment (27%, p = 0.0318), accompanied by a 33% reduction in 6E10-Aβ plaque (p = 0.0066; Fig. 2n). Consistent with our findings in AD patients, elevated cerebral H3K9me3 levels in AD-model mice strongly associated with cognitive deficits and neuropathology (Fig 2o and Extended Data Fig. 5d). On day 4, higher cortical H3K9me3 levels correlated with poorer performance in the colour- and contrast-mode X-maze ( r = −0.77, p< 0.0001 and r = −0.48, p= 0.0416, respectively). On day 15, elevated cortical and hippocampal H3K9me3 levels correlated with increased errors in the Barnes maze (CC: retention, r = 0.77, p= 0.0053; Hipp: reversal, r = 0.57, p= 0.0113 and retention, r = 0.49, p= 0.0299) and reduced freezing in the fear conditioning (Hipp: r = −0.63, p= 0.0063). Notably, H3K9me3 levels strongly associated with Aβ plaque burden in both cortex (day 4: r = 0.79, p= 0.0020; day 15: r = 0.84, p= 0.0011) and hippocampus (day 15: r = 0.82, p= 0.0033), and with astrogliosis (day 4: r = 0.72, p= 0.0002) in the cortex. Collectively, these findings show that a single dose of ETP69 drives a sustained reduction of cerebral H3K9me3, astrogliosis, and Aβ pathology in aged AD-model mice. Since dendritic spine loss is a strong predictor of cognitive decline 50-52 , we asked whether cerebral H3K9me3 reduction preserves synaptic architecture in the context of AD. We therefore quantified dendritic spine structure and integrity by Golgi–Cox staining in brains from aged ETP69-treated AD + and WT mice (regimen B; Fig. 3a,b). Analysis of spine subtypes, showed reduced thin spine density (thin spines/μm of dendrite), reflecting newly formed synapses, in control AD + mice, with 28% reduction in the cortex (p = 0.0004) and 14% in the hippocampus (p = 0.0327; Fig. 3e). ETP69 substantially increased their density in both regions and genotypes (WT: CC, 1.2-fold, p = 0.0006; Hipp, 1.3-fold, p = 0.0013; AD + :CC, 1.6-fold, p < 0.0001; Hipp, 1.5-fold, p = 0.0003; Fig. 3e). In hippocampus, ETP69 also increased stubby spine density (2-fold, p = 0.0461; Fig. 3e) and reduced mushroom spines (54%, p = 0.0085; Extended Data Fig. 6c), with no significant change in the cortex. The overall spine number increased with ETP69 in both brain regions and genotypes (Fig. 3f). Importantly, increased thin spine density correlated with improved performance in the Barnes maze and contextual fear conditioning ( r = −0.54, p = 0.0301 and r = 0.67, p = 0.0063 respectively; Fig. 3g). By quantifying spine subtype composition as a proportion of total spines, we found that thin spines were the predominant subtype in aged WT mice (60–70%) and remained dominant but were reduced in AD + mice (~50%; Fig. 3h and Extended Data Fig. 6d). ETP69 increased the thin-to-total spine ratio while decreasing the mushroom-to-total spine ratio (Extended Data Fig. 6e). Notably, Barnes maze performance inversely correlated with the thin-to-total spine ratio (r = −0.56, p = 0.0232; Fig. 3i). Finally, hippocampal synaptic proteins were increased in ETP69-treated AD⁺ mice, with higher postsynaptic density protein PSD95 (1.3-fold, p = 0.0137; Fig. 3j) and presynaptic synaptophysin SYP (1.9-fold, p = 0.0140; Fig. 3k). Together, these data suggest that lowering cerebral H3K9me3 restores synaptic integrity, underpinning epigenetic-based cognitive recovery in aged AD-model mice. Targeting H3K9me3 attenuates tauopathy and amyloidosis, reshaping microglia and monocyte responses in midlife AD models In the next set of experiments, we examined the effects of H3K9me3 reduction on tau and amyloid pathology, as well as cell-type–specific responses, in 14-month-old AD transgenic mice harbouring established amyloidosis and/or tauopathy, before the onset of ageing processes. In 3xTg-AD mice (APPSwe/tauP301L/PS1 tm1Mpm ), which develop hyperphosphorylated tau and NFTs by the age of 12 months 53 , a single intraperitoneal injection of ETP69 (10 mgkg -1 ) significantly reduced cortical levels of H3K9me3 and paired helical filaments (PHF)-tau IR (62%, p = 0.0064 and 39%, p = 0.0003, respectively; Fig. 5a,b). Furthermore, elevated cortical H3K9me3 strongly correlated with PHF-tau accumulation ( r = 0.78, p = 0.0050; Fig. 5c). In 14-month-old APPswe/PS1dE9 (AD⁺) mice, a single intraperitoneal injection of ETP69 (10 mgkg -1 ) restored cognitive functions as observed in the colour X-maze and fear conditioning (Fig. 4d–f and Extended Data Fig. 7a–d). While locomotor activity, measured by total arm entries, was unaffected by ETP69 (Extended Data Fig. 7a), the cognitive deficits observed in AD + versus WT mice (X-maze: p = 0.0009; Fear conditioning: p = 0.0021) was reversed by ETP69, as shown by increased percentage of X-maze alternations (p = 0.0089; Fig. 4e) and prolonged freezing in the fear conditioning (p = 0.0203; Fig. 4f). ETP69 also enhanced cognitive performance in WT mice (Extended Data Fig. 7b–c). In contrast, AD-associated visual impairment, manifested by reduced B↔W bidirectional transitions (p = 0.0278), was not significantly rescued by the single ETP69 dose (Extended Data Fig. 7d). To determine whether cognitive improvement at this earlier disease stage was accompanied by underlying tissue and cellular remodelling, we performed complementary neuropathological analyses. Indeed, AD + mice exhibited elevated cortical H3K9me3 relative to WT controls (1.6-fold; p < 0.0001), and ETP69 treatment normalized H3K9me3 in AD + mice while further reducing H3K9me3 in WT mice (40–42% decrease; p < 0.0001; Fig. 4g and Extended Data Fig. 7e). Consistent with our findings in aged mice, ETP69 mitigated neuropathology in these younger AD + animal models, significantly reducing 6E10-Aβ plaque burden (40%, p = 0.0019; Fig. 4g) and GFAP astrogliosis (48%, p < 0.0001; Fig. 4g and Extended Data Fig. 7e,f). Pearson’s r correlation analyses further showed that lower cortical H3K9me3 was associated with improved cognitive performance as detected in the fear conditioning ( r = −0.59, p = 0.0037), and colour-mode X-maze ( r = −0.59, p = 0.0061), as well as with reduced Aβ plaque load ( r = 0.78, p = 0.0042; Extended Data Fig. 7g). To define the cell-type–specific effects of ETP69 on H3K9me3, we performed co-immunolabelling for H3K9me3 with neuronal (NeuN; Fig. 4h), astrocytic (GFAP; Fig. 4i), and myeloid (IBA1 + CD45 + microglia and monocyte/macrophages; Fig. 4j) markers in the brains of these AD + murine models. H3K9me3 was preferentially lowered in myelomonocytes (69%, p = 0.0002) and neurons (45%, p < 0.0001), whereas astrocytes showed only a non-significant trend (18%, p = 0.0695). Building on this myeloid cell selectivity, we next interrogated the inflammatory milieu surrounding Aβ plaques. ETP69 treatment markedly reduced cortical IBA1 immunoreactivity (40% reduction; p = 0.0021; Fig. 4k and Extended Data Fig. 7h), and the extent of cortical H3K9me3 lowering closely correlated with diminished microgliosis (r = 0.69; p = 0.0182; Fig. 4l), linking H3K9me3 derepression to inflammatory response. Notably, ETP69 enhanced infiltration of monocyte‑derived macrophages (IBA1⁺CD45 high ) surrounding 6E10-Aβ plaques by 1.8‑fold (p = 0.0006; Fig. 4m and Extended Data Fig. 7i). In parallel with reduced plaque burden and microgliosis, ETP69 robustly shifted the cortical myeloid compartment towards peripheral infiltration, increasing the proportion of IBA1 + CD45 high peripherally derived macrophages within total IBA1 + cells by 3.2-fold compared to control AD + mice (p = 0.0004; Fig. 4m). The expanded brain infiltrating monocyte/macrophage‑covered area inversely correlated with cortical Aβ load (r = −0.69, p = 0.0173; Fig. 4n), consistent with their reported phagocytic role in Aβ clearance 54-61 . Peripheral immune profiling in a subset of 18‑month‑old AD⁺ mice further supported this mechanism: ETP69 markedly increased circulating monocytes (3.8‑fold, p = 0.0039) and granulocytes (2.9‑fold, p = 0.0375), without altering lymphocyte or red blood cell counts (Fig. 4o and Extended Data Fig. 7j). Together, these findings indicate that ETP69 preferentially reduces cerebral H3K9me3 within microglia and peripherally derived mononuclear phagocytes, and promotes their accumulation at Aβ plaque sites, thereby enhancing Aβ clearance and limiting neuroinflammation. Limiting H3K9me3 re-establishes brain proteostasis and strengthens immune pathways supporting Aβ clearance To delineate the molecular basis of neuroprotection conferred by SUV39H1 inhibition–mediated H3K9me3 lowering, we performed global MS-based proteomic profiling of brains from the same mouse cohort. Among 5097 proteins identified across all samples (Fig. 5a and Extended Data Fig. 8a), 4358 were detected in at least three animals per group, and 1119 DEPs (p < 0.05) distributed between the following three comparisons: AD + versus WT (AD + /WT: 747 DEPs), ETP69- versus vehicle-treated WT (WT ETP69/DMSO: 318 DEPs), and ETP69- versus vehicle-treated AD + mice (AD + ETP69/DMSO: 371 DEPs). Only 40 DEPs overlapped between WT ETP69/DMSO and AD + ETP69/DMSO, indicating genotype-specific proteomic responses to treatment consistent with distinct H3K9me3 landscape in healthy versus AD + brains (Fig. 5a). Overall, ETP69 elicited a balanced distribution of down- and upregulated DEPs; however, applying a >1.2-fold-change threshold revealed a pronounced shift toward upregulation in both genotypes (Fig. 5b; DEPs are listed in Extended Data Tables 2–7). Heatmaps and principal component analysis (PCA) showed clear clustering by genotype and treatment group (Fig. 5c,d and Extended Data Fig. 8b,c). Notably, of the 162 DEPs shared between the AD + /WT and AD + ETP69/DMSO comparisons, ETP69 reversed 159 AD-associated changes (R 2 = 0.86, p 1.04), 1683 of 2005 shared proteins (84%) were similarly redirected toward WT-like levels (Extended Data Fig. 9). Gene Ontology enrichment analysis of these treatment-driven reversals revealed coordinated upregulation of pathways involved in cytoskeleton organization, synaptic function, and dendritic spine maintenance, accompanied by downregulation of AD-linked metabolic and lipid oxidation. PANTHER classification showed that metabolite-converting/protein-modifying enzymes, nucleic acid metabolism and translation, and membrane trafficking/transport were the most represented functional categories across comparisons (Extended Data Fig. 10a). Volcano plots highlight the top 20 up- and top 20 downregulated DEPs in the treatment versus control AD + brains (Fig. 5f; AD + /WT and WT ETP69/DMSO comparisons in Extended Data Fig. 10b). To define the molecular networks targeted by lowering cerebral H3K9me3 in AD models, we next performed IPA. As expected, IPA predicted significant activation of the amyloid precursor protein (APP) in AD + compared to WT mice (z-score = 2.39, adj. p < 0.0090, Fig. 5g), including canonical AD markers, such as clusterin (CLU), apolipoprotein E (APOE), midkine (MDK), modulates APP processing and Aβ generation and the microglial activation marker galectin-3 (LGALS3); yet their expression was unchanged by this epigenetic-targeting drug. Nevertheless, 19 additional APP-linked proteins, including metallothionein 1 (MT1), lysosome-associated membrane glycoprotein 1 (LAMP1), ketimine reductase μ-crystallin (CRYM), microtubule-associated protein 6 (MAP6), ATP-binding cassette sub-family G member 1 (ABCG1), prostaglandin-H2 D-isomerase (PTGDS) and γ-aminobutyric acid type B receptor subunit 2 (GABBR2), shifted toward WT-like abundance in AD + brains after treatment, consistent with selective normalization of AD-associated proteomic dysregulation. Interestingly, microtubule-associated protein tau (MAPT), a major component of the neuronal cytoskeleton, was also a top downregulated protein in AD + versus WT mice and strongly upregulated in response to ETP69 in AD + mice (2.4-fold, p = 0.0188; Fig. 5g). Notably, the stress- and inflammation-responsive proteins MT1 and MT2 were among the most strongly downregulated DEPs in ETP69-treated AD + mice (Fig. 5f,h). Reduced MT2 abundance by ETP69 was independently validated by IHC in aged AD + mice (Fig. 5i). In addition, the neuroprotective haemoglobin-scavenging proteins haptoglobin (HP) and hemopexin (HPX) were among the most strongly upregulated proteins in ETP69-treated WT and AD + mice (Fig. 5j). HPX was reduced in AD + versus WT mice, revealing a disease-associated deficit that ETP69 restored. Consistently, circulating HPX and HP increased markedly after treatment and strongly correlated with their cerebral abundance (Fig. 5k and Extended Data Fig. 11a,b), indicating systemic engagement of an early innate immune response. In parallel to our blood and brain findings, proteomics from the same cohort showed that ETP69 enhanced pathways involved in Aβ clearance and innate immune activation (Fig. 5l,m and Extended Data Fig. 11c–e). Intercellular adhesion molecule 1 (ICAM1) and guanine nucleotide exchange factor 3 (VAV3), which facilitate cerebral monocyte recruitment and transendothelial migration, were induced in ETP69-treated brains. Additionally, the low affinity immunoglobulin gamma Fc region receptor II (FCGR2B), Rho GTPase-activating protein 25 (AHRGAP25), and engulfment and cell motility protein 1 (ELMO1), key regulators of the phagocytic process, were upregulated following treatment. Complement components CFH and C3 increased substantially, and brain C3 correlated with circulating C3 (Fig. 5m and Extended Data Fig. 11d,e), consistent with systemic-to-CNS innate immune priming. The rise in C3, implicated in microglial Aβ clearance, coincided with reduced cytochrome P450 family 51 (CYP51) and LRP receptor-related protein-associated protein 1 (LRPAP1), two negative regulators of microglia-dependent Aβ removal (Fig. 5m). Notably, ICAM1, FCGR2B and CFH, already elevated in AD + mice, were further increased by ETP69, suggesting that H3K9me3 inhibition amplifies an existing neuroimmune response to enhance microglia- and monocyte-mediated chemotaxis and phagocytosis, thereby promoting more efficient cerebral Aβ clearance. Together, these proteomic signatures indicate that lowering cerebral H3K9me3 broadly resets AD brains toward a WT-like state by coordinately restoring synaptic–cytoskeletal programs and potentiating complement-linked phagocytic immunity to support Aβ clearance. Restored neurotrophic VGF–granin signalling tracks cognition We next examined whether H3K9me3 inhibition engages molecular programs linked to neuroprotection and behaviour. IPA predicted significant activation of pathways and upstream regulators related to neuroplasticity as well as learning and memory, reversing the deficits observed in AD + mice (Fig. 6a-c and Extended Data Fig. 12a–c); a STRING network highlighted shared DEPs across these functions (Extended Data Fig. 12d). Among these, leucine-rich repeat neuronal protein 4 (LRRN4), a regulator of hippocampus-dependent learning and long-lasting memory, was strongly induced in ETP69-treated versus control AD + brains (1.73-fold, p = 0.015). Five proteins were common to all three IPA functions—NTRK2 (TRKB), SHANK3, SPAST, TSC2 and NGF-inducible neurosecretory protein (VGF; Extended Data Fig. 12d). Notably, IPA identified RICTOR, GABA and BDNF activation by ETP69 treatment (z scores: 3.59, 2.50, 2.45, respectively), whereas these regulators were among the most strongly inhibited in AD + mice (Fig. 6c). Given the key role of BDNF in neuronal survival, synaptic plasticity, learning, and memory, we further analyzed downstream targets in the BDNF network (Fig. 6d). The prediction of BDNF activation was based on the downregulation of phospholipid transfer protein (PLTP) and annexin A2 (ANXA2), which was further validated by hippocampal analysis of aged AD + mice treated with ETP69 (Fig. 6e). Importantly, BDNF activation network identified two proteins at the intersection of learning, dendritic spine maintenance, and neuronal survival, VGF and neurotrophic receptor tyrosine kinase 2 (NTRK2; a BDNF receptor; Fig. 6d and Extended Data Fig. 12d). Indeed, ETP69 induced cerebral NTRK2 in both WT and AD + mice alongside upregulating the proprotein convertase subtilisin/kexin type 1 (PCSK1) in AD + mice (Fig. 6f); the latter is a convertase that cleaves VGF into TLQP-62 and TLQP-21 active neuropeptides. Given VGF’s membership in the granin family, we assessed granin-related proteins and found that VGF, CHGA, CHGB/SCG1, SCG2, SCG5 (7B2) and PCSK1N/SCG8 were reduced in AD + versus WT brains and shifted upward to WT-like levels after ETP69; SCG3 showed the opposite pattern and was similarly normalized (Fig. 6g). We further validated BDNF and VGF expression in two independent cohorts using WB and IHC (Fig. 6h−j). H3K9me3 inhibition led to hippocampal BDNF induction (WB: 1.3-fold, p = 0.0284; IHC: 1.3-fold, p = 0.0599), whereas VGF increased robustly in both aged WT (WB: 1.3-fold, p = 0.0017; IHC: 1.4-fold, p = 0.0215) and AD + mice (IHC: 1.9-fold, p < 0.0001; WB: 1.4-fold, p = 0.0004), with prominent signal in the dentate gyrus subgranular zone (Fig. 6j). Cortical VGF was reduced in AD + versus WT mice (32%, p = 0.0003) and was restored by ETP69 (1.9-fold, p = 0.0075). Alongside decreased H3K9me3 (28%, p = 0.0069), amyloid plaques were decreased in the hippocampus (12F4-Aβ: 27%, p = 0.0015; ThioS: 38%, p = 0.0333) and cortex (ThioS: 17%, p = 0.0080) following ETP treatment (Fig. 6h−j). Importantly, cerebral VGF inversely correlated with H3K9me3 in the hippocampus (IHC: r = −0.64, p = 0.0022; WB: r = −0.60, p = 0.0412) and cortex (r = −0.68, p = 0.0214), and associated with reduced Barnes maze errors reflecting improved memory retention (r = −0.58, p = 0.0088; Fig. 6k), highlighting VGF as a direct mechanistic bridge from epigenetic modulation to cognitive benefit. Retinal immune–synaptic proteostasis and colour vision rescue by SUV39H1 inhibition To explore the impact of H3K9me3 inhibition on the retina of AD + mice, we studied the MS-based proteome profiles of ETP69-treated 14-month-old WT and AD + cohort mice (Fig. 7 and Extended Data Figs. 13-14). We detected 5,107 proteins in ≥3 animals per group and identified 732 DEPs (p < 0.05) across AD + /WT (226), WT ETP69/DMSO (452) and AD + ETP69/DMSO (153) comparisons (Fig. 7a). As in brain, WT and AD + treatment responses minimally overlapped (11 DEPs), and brain–retina concordance was limited (Extended Data Fig. 13a; only three shared DEPs in either WT ETP69/DMSO or AD + ETP69/DMSO), indicating genotype- and tissue-specific drug effects. Heatmaps and PCA segregated samples by genotype and treatment (Fig. 7b and Extended Data Fig. 13b,c). Notably, all 49 DEPs shared between AD + /WT and AD + ETP69/DMSO were fully reversed by treatment in AD + retina (Fig. 7c). Volcano plots highlight the top 20 up- and downregulated DEPs in treated versus control AD + retina (Fig. 7d; AD + /WT and WT ETP69/DMSO in Extended Data Fig. 14a; DEPs with |FC| > 1.20 in Extended Data Tables 8–13). GO enrichment across 330 DEPs (177 AD + /WT, 104 AD + ETP69/DMSO and 49 shared) identified AD-driven perturbations, and ETP69 caused restoration, of apoptosis, synapse organization/dendritic spine and neurotransmission, and acute immune response pathways (Fig. 7e,f). Consistent with our brain and blood datasets, HP and HPX were among the top induced retinal DEPs in ETP69-treated AD + mice (HP FC = 3.28, p = 0.0084; HPX FC = 2.18, p = 0.0319), with similar but non-significant trends in WT (HP FC = 1.63, p = 0.0997; HPX FC = 1.30, p = 0.1315; Extended Data Table 12). Orosomucoid 2 (ORM2) was likewise strongly upregulated in both genotypes (Fig. 7d and Extended Data Fig. 14a). In AD + retina, ETP69 reduced pro-inflammatory coagulation factor III (F3) and thrombin (F2), an IPA-predicted upstream regulator (Fig. 7f and Extended Data Fig. 14b). IPA further predicted activation of TNF and SP1 in AD + retina (z = 2.43 and 2.38), with ETP69 fully reversing TNF (z = −2.02) and partially attenuating SP1 (z = −0.97; Fig. 7f and Extended Data Fig. 14b), consistent with modulation of retinal Aβ-, apoptosis- and inflammation-linked signalling. All DEPs with histone-modifying activity in AD + retina were reversed by ETP69 (Fig. 7e). Specifically, RIOX1, PPP6C and the retina-enriched kinase STK35 (reduced in AD + retina) were increased by treatment, whereas the H3K9 deacetylase SIRT1 (elevated in AD + retina) was decreased. ETP69 also increased RSBN1 (H4K20 demethylase activity) while reducing CLOCK and GSK3A in AD + retina. Importantly, IPA of direct/indirect neighbour DEPs predicted SUV39H1 activation in AD + retina (Extended Data Fig. 14c) and its inhibition after ETP69 (Fig. 7g); SUV39H1, SIRT1, CLOCK and GSK3A are established modulators of circadian rhythm, ageing and retinal disease. Synapse- and neurotransmission-related pathways comprised a major component of AD + retinal dysregulation and were broadly restored by ETP69 (Fig. 7e). PICK1, which regulates AMPAR internalization and synaptic plasticity, was reduced by treatment (and elevated in AD + retina), whereas ADGRL3 (postsynaptic spine-enriched) and CRIPT (anchors PSD95 at excitatory synapses) were increased in treated AD + retina (Extended Data Fig. 14b), consistent with improved synapse-supportive programs. In parallel to the brain, ETP69 also modulated extended granin family members in AD + retina: CHGA and PCSK1N were upregulated in treated AD + retina (with PCSK1N also increased in WT), whereas VGF was uniquely reversed and downregulated by ETP69 in AD + retina—opposite to brain (Fig. 7h), consistent with reports of elevated VGF in retinal neurodegeneration and its prominent expression in Müller/astrocytic glia. GO terms further implicated visual behaviour and learning (Fig. 7e). In colour-mode X-maze testing, WT alternations preferentially followed continuous W↔R↔G↔B sequences, a pattern reduced in AD + mice but restored by ETP69 (Fig. 7i); these sequences typically began or ended at the white arm, efficiently avoiding bidirectional B↔W transitions (Fig. 7i). Together, these data show that H3K9me3 inhibition selectively reprograms the AD + retinal proteome toward WT-like synaptic and immune homeostasis, restoring visual learning. Discussion In this study, we identify the repressive heterochromatin mark H3K9me3 as an early, cross-compartment epigenetic signature of AD across the brain–retina axis, consistent with a shift toward chromatin compaction that constrains neuronal and immune programs. In matched human donor brain and retinal tissues, H3K9me3 was already elevated at the MCI stage, tracked neuropathological burden and cognitive dysfunction, and showed strong brain–retina concordance. Proteomic profiling of AD cortex and retina indicated enrichment of pathways governing heterochromatin formation and histone modifications, predicting SUV39H1 activation and H3K9 hypertrimethylation. Mechanistically, our interventional studies in transgenic AD models revealed that pharmacologic inhibition of SUV39H1 rapidly and durably lowers cerebral H3K9me3, recalibrates brain and retina neuroimmune competence and synaptic proteostasis, attenuates gliosis and amyloid/tau pathology, culminating in rescue of dendritic spine architecture and visuo-cognitive function. These benefits are attributed, at least in part, to a tissue-specific restoration of a trophic–immune axis, centered on neurosecretory VGF and the extended granin family, bridging epigenetic remodeling with functional recovery. Together, these observations support a model in which age- and disease-associated epigenomic regulation at the brain–retina interface can actively stabilize maladaptive cellular states rather than simply reflect downstream pathology. The synchronized H3K9me3 abundance between the dorsolateral prefrontal cortex and the superotemporal retina across cognitively normal individuals and patients spanning the AD continuum positions the retina as an accessible CNS readout of epigenetic mark of disease severity. In both compartments, H3K9me3 rises from the MCI stage and further in AD, tightly correlating with amyloid and tau proteinopathy, and tracking disease stage (ABC, Braak) and cognitive impairment (CDR, MMSE). These observations are consistent with prior reports of elevated H3K9me3 in the hippocampus and in orbitofrontal and temporal cortices of individuals with AD 39,62,63 . The direct correlation between retinal and prefrontal cortical H3K9me3 is in line with mounting evidence that the retina recapitulates key molecular and pathological features of AD 13,14,16-20,25,27,64-68 . Proteomics in human brain and retina tissues converge on chromatin-related dysregulation consistent with enhanced H3K9me3, yet tissue-specific: in the brain, altered PRMT1/PRMT5 expression and prediction of SUV39H1 activation supported increased H3K9me3, while in the retina, downregulation of KDM4B/C and KAT7 together with increased JMJD6 indicated the same shift. Although the specific molecular players differ between tissues, consistent with their distinct cellular compositions, the directionality of epigenetic remodelling is preserved, indicating a conserved disease mechanism rather than coincidental tissue-specific changes. Future studies should delineate which epigenetic biomarkers are shared across compartments and which reflect tissue-specific regulatory programs in healthy aging and disease. Building on these human data, we demonstrate a causal contribution of excessive H3K9me3 in AD-like phenotypes and that targeting H3K9me3-dependent repression is disease modifying in preclinical AD models. Our studies provide the first evidence that pharmacological SUV39H1-H3K9me3 inhibition (ETP69) in midlife and aged AD mouse models (APPswe/PS1dE9 and 3xTg-AD) attenuates amyloidosis and tau pathology, dampens neuroinflammation, and restores synaptic integrity and cognitive performance. Prior work also showed the effect of ETP69 in aged WT mice 69 . In our study, lowering H3K9me3 in AD-model mice reactivated cerebral and retinal molecular pathways governing synaptic plasticity, dendritic spine dynamics, neuronal survival, and learning, while simultaneously reshaping the neuroimmune environment toward enhanced amyloid clearance and reduced gliosis. Dendritic spine remodelling is central to learning and memory 70,71 , and our proteomics data predicted increases in spine density and branching by H3K9me3 inhibition. Structural analyses by Golgi-Cox confirmed the emergence of new filopodia and long-thin spines—critical for encoding new memories 72 —in ETP69-treated AD-model mice to levels comparable to or exceeding WT, providing a morphological correlate for cognitive recovery. At the molecular level, IPA predicted activation of the BDNF network, with marked upregulation of BDNF receptor NTRK2 in AD + mice, exceeding WT levels, following treatment. Given that Bdnf transcription is epigenetically regulated 73 and sensitive to H3K9 di- and tri-methylation 69,74,75 , and that CREB–BDNF–NTRK2 signalling is diminished in AD 76-79 , restoration of this pathway provides mechanistic link to cognitive rescue. A key BDNF effector 80 , VGF was increased in brains of ETP69-treated AD + mice to levels equal or higher to that of WT mice, and inversely correlated with H3K9me3 levels. Harmonizome data identify Vgf among the 401 genes with H3K9me3-rich promoters 81 , suggesting direct epigenetic derepression of Vgf upon H3K9me3 reduction. The protease PCSK1 that cleaves VGF into bioactive peptides TLQP-62 and TLQP-21 82 , was upregulated alongside VGF. TLQP-62 has been implicated in dendritic branching, synaptic plasticity and long-term memory 80,83-85 , and its reported roles in neurogenesis 86,87 are consistent with the prominent VGF induction we observed in the dentate gyrus subgranular zone. Since TLQP-62 activates VGF translation via mTOR signalling 88 and requires NTRK2 for hippocampal memory formation 80 , our findings support an mTOR-mediated, positive PCSK1–VGF–NTRK2 regulatory loop that reinforces the cognitive benefits of H3K9me3 inhibition. Other members of the extended granin family 89 , including CHGA, CHGB, SCG2, 7B2, and PCSK1N, were upregulated following treatment. Multiple omics studies have shown downregulation of cerebral VGF along with, albeit to a lesser extent, all other granin family members (CHGA, CHGB, SCG2, SCG3, 7B2 and PCSK1N) in AD patients 90-104 as well as in a murine AD model 91,100 . As granins regulate vesicles biogenesis and neuropeptide release critical for neuronal activity and cognitive function 105-108 , their restoration likely contributes to improved synaptic transmission. Beyond synaptic function, granins and VGF-derived peptides intersect with innate immunity and Aβ processing and clearance. Granin family members localize near Aβ plaques in brains of AD patients and mouse models, exerting neuroprotective effect; 7B2 and PCSK1N inhibit amyloid aggregation 121,122 and efficiently prevent Aβ neurotoxicity. VGF-derived peptide TLQP-21 enhances microglial phagocytosis and chemotaxis via C3AR1 to reduce Aβ plaque load and dystrophic neurites 123-125 , consistent with reduced cerebral Aβ burden and a dystrophic neurite marker LAMP1 in ETP69-treated AD-model mice. Treatment further increased circulating and brain-infiltrating monocytes and elevated innate immune mediators, including haptoglobin, hemopexin and complement C3; in parallel, H3K9me3 targeting increased cerebral ICAM1 with its receptor ITGAM/ITGB2 109 , which also binds C3 110 , suggesting augmented monocyte recruitment and phagocytic capacity, in line with evidence that peripheral monocyte enrichment promotes Aβ clearance and preserves cognition 54-61,111-120 . Collectively, these data indicate that H3K9me3 acts as a trophic–immune checkpoint inadequately engaged in AD, and its inhibition enhances the brain’s capacity to counter amyloidosis and chronic neuroinflammation. In the retina, ETP69 robustly activated synaptic and immune programs, yet treatment-responsive DEPs showed minimal overlap with the brain, consistent with tissue-specific proteomic networks. VGF regulation was notably divergent, increased in AD + retina but decreased in AD + brain, yet in both compartments it shifted toward WT-like levels with treatment, highlighting the decisive influence of local cell-type context on epigenetic-to-proteomic outputs; in the retina, VGF is predominantly expressed by Müller glia and astrocytes, whereas in the brain it is primarily neuronal. Like the brain, retinal granins including CHGA and PCSK1N were elevated by ETP69. Although acute-phase proteins HP and HPX were induced across brain, retina and blood, treatment elicited a retina-selective, disease-corrective proteomic response that opposed AD-associated apoptosis and TNF immune activation leading to rescue of synapse and colour vision. Our study integrates cross-compartment human datasets with mechanistic, interventional validation in midlife and aged AD mouse models, yet several considerations remain. First, the human analyses are cross-sectional and correlative, limited by sample size and heterogeneity; larger, longitudinal studies will be needed to determine the temporal dynamics of H3K9me3 and its relationship to disease onset and progression in humans. Second, the reproducible molecular and functional rescue elicited by SUV39H1–H3K9me3 inhibition across AD models motivates deeper mechanistic resolution, including cell-type–resolved epigenomic interrogation using chromatin accessibility and single-cell transcriptomic approaches. Third, translation will require safety and efficacy testing of H3K9me3-targeted treatment in AD patients, alongside development and clinical validation of noninvasive retinal readouts for specific chromatin marks to establish the retina as a proxy for cerebral epigenetic state. In conclusion, this study defines H3K9 hypertrimethylation as an epigenetic synaptic–immune checkpoint linked to neurodegeneration in AD. This repressive heterochromatin signature spans brain and retina in human AD and preclinical models. Pharmacologic SUV39H1–H3K9me3 inhibition reconfigures chromatin to restore synaptic plasticity and immune competence programs, culminating in recovery of cognitive and visual function. By coordinately resetting multiple dysregulated pathways, epigenomic modulation emerges as a transformative alternative to therapies that solely target proteotoxic clearance. These findings resonate with growing interest in epigenetic reprogramming to counteract ageing and neurodegeneration, including ocular ageing where epigenetic drift contributes to visual decline 126,127 . The brain–retina concordance further positions the retina as a noninvasive surrogate for epigenetic biomarker imaging to guide early precision diagnosis and therapy. Methods Mice Double-transgenic B6.Cg-Tg (APPswe/PSEN1dE9) 85Dbo/Mmjax mice, a mouse model of Alzheimer’s disease (AD + ), and their age-matched WT C57BL/6J littermates were obtained from the Mutant Mouse Resource and Research Center (MMRRC) at the Jackson Laboratory (RRID:MMRRC_034832-JAX). Mice were bred and maintained at Cedars-Sinai Medical Center vivarium under standardized conditions: housed up to five per cage on a 14-hour light/10-hour dark cycle, ambient temperature maintained at 74 °F (23 °C) ± 2 °F, and relative humidity at 30–70%, with ad libitum access to food and water and a maximum of five animals per cage. All AD + animals used in this study had a congenic C57BL/6 background. Both male and female mice were used for all experiments and were assigned to experimental groups after balancing for age and genotype. A cohort of female triple-transgenic B6;129-Tg (APPswe/tauP301L)1Lfa Psen1 tm1Mpm (3xTg AD + ) mice (Jackson Laboratories, MMRRC_034830-JAX), maintained at the University of California, Irvine (UCI), was included for a limited subset of experiments and analysed separately where indicated. All procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Cedars-Sinai Medical Center and UCI and conducted in accordance with the NIH and ARRIVE Guidelines for the Care and Use of Laboratory Animals. ETP69 treatment regimen Lyophilized ETP69 was resuspended in 50% DMSO (in saline) at a concentration of 10 mg/ml. ETP69-treated mice received intraperitoneal injections of ETP69 (10 mg/kg), and control mice (DMSO-injected groups) received vehicle injections (50% DMSO in saline). Four cohorts of WT and AD + mice (n=153; 14 and 18 months old) were treated with ETP69 or DMSO under three regimens (regimens S, R, and B), with behavioural timelines and experimental endpoints shown in Fig. 2b (18-month-old cohort) and Fig. 4d (14-month-old cohort). Two cohorts of AD + and WT mice received a single ETP69 dose (regimen S) 1 day before the start of behavioural testing (day 0) and were euthanized either on day 4 or day 15. A third group of AD + and WT mice received repeated ETP69 doses (once weekly for 11 weeks) with the last administration on day 0 (regimen R). A fourth group of AD + and WT mice received the first dose of ETP69 on day 0 and a booster dose on day 9, the day before memory retention testing of the Barnes maze (regimen B). Control AD + and WT mice received DMSO according to the same regimens S, R, or B. One cohort of female 3xTg AD + mice (n=11; 14 months old) received a single injection of ETP69 or DMSO (Fig. 4a). Behavioural tests All WT and AD + animals in this study underwent behavioural tests. Open field test Spontaneous locomotor activity was assessed for 30 min using the Photobeam Activity System (www.sandiegoinstruments.com). Ambulatory and rearing activities were recorded, and distance, speed, and resting time were calculated. Barnes maze test (hippocampus-based spatial learning and memory test) Mice were first trained to locate an escape box in a 20-hole circular table during a 4-min trial that was performed 3 times per day for 4 days (acquisition training phase), as previously described 57 . Following a 2-day break, the memory retention of each mouse was evaluated on day 7 (retention phase). Memory extinction and learning of a new escape location were assessed on days 8–9 (reversal phase). The latency to find the escape box and the number of incorrect entries (errors) were recorded for each trial and averaged on each day for each mouse. Y-maze spontaneous alternation test (hippocampus-based spatial working memory test) The Y-shaped apparatus used for this study consists of two arms equal in length and one longer arm. Mice were individually placed at the distal end of the long arm and allowed to move freely throughout the entire maze (all three arms) for 5 min under dim light. The sequence of arm entries and the total number of entries were recorded; the percentage of spontaneous alternations was calculated as follows: A spontaneous alternation is defined as the sequential visit to the three different arms without returning to a previously visited one. Visual-stimuli X-maze test (visual-cognitive test) To assess spontaneous behaviour induced by colour (under equal conditions) and contrast sensitivity, the mice were tested using our custom-made colour and contrast-mode X-maze, as previously described 27,128,129 . For each mode, the mice were individually placed in the centre of the ViS4M and allowed to freely explore the maze for 5 min. The sequences of arm entries were manually documented according to the video recordings. The total number of entries, percentage of bidirectional transitions between arms, and percentage of alternations were all determined from the sequences of arm entries. Chord diagrams were generated to visualize behavioural data from the Barnes maze and X-maze tests using the free online resource Circos (mkweb.bcgsc.ca/tableviewer/) as previously described 128,129 . Context-specific fear conditioning test (spatial associative memory test) During the acquisition phase, an animal was placed in a freezing behaviour-monitoring chamber (www.sandiegoinstruments.com) and allowed to habituate for 2 min before receiving a 0.2-mA electric foot shock for 1 s. The animal remained in the chamber for an additional 3 min and was then returned to its home cage. To assess context-specific fear, the animal was placed in the same chamber 24 h after the acquisition phase, and the freezing time (absence of movement for 3 s) over a 4-min session was recorded. All behavioural tests were performed by an experimenter blinded to mouse genotypes and treatments. Mouse brain collection and processing After completion of the behavioural tests, all mice were deeply anaesthetized (50 mg/kg ketamine/xylazine) and transcardially perfused with ice-cold saline solution containing 0.5 mM EDTA. Mouse brains were collected and processed as follows: 1) snap frozen and stored at −80°C for protein extraction; 2) fixed in 2.5% PFA overnight and then cryoprotected in 30% sucrose for immunohistological analyses; or 3) processed for Golgi-Cox staining. Fixed brains were coronally sectioned at 30 µm thickness using a cryostat (Leica CM3050 S; Leica Biosystems, Nussloch, Germany). Sections were stored at 4°C in PBS containing 0.01% sodium azide in 24-well plates, until immunochemical processing. Postmortem human brains and retinas Human brain and retinal tissues were obtained from the Alzheimer’s Disease Research Center (ADRC) Neuropathology Core in the Department of Pathology (IRB protocol HS-042071) of Keck School of Medicine at the University of Southern California (USC, Los Angeles). USC-ADRC maintains human tissue collection protocols that are approved by institutional managerial committees and subject to oversight by the National Institutes of Health. The ADRC provided clinical and neuropathological reports on patients’ neurological examinations, neuropsychological and cognitive tests, family history, and medication lists, as collected in the ADRC system using the Uniform Data Set (UDS). Histological studies were performed under an approved IRB protocol at Cedars-Sinai Medical Center. We examined brains and retinas from deceased patient donors with clinically and neuropathologically confirmed AD (n=13) or mild cognitive impairment (MCI due to AD, n=6) as well as brains from deceased individuals with normal cognition (CN [control], n=6). Donor information is provided in Extended Data Table 1. Fresh brain tissues (frontal cortex) were snap frozen and stored at −80°C. Portions of fresh-frozen brain tissues were fixed in 4% PFA for 16 h and then dehydrated in 30% sucrose/PBS. The brain tissues were coronally sectioned (30 μm thick) on a cryostat (Leica CM 3050_S) and mounted on slides coated with 3-aminopropyltriethoxysilane (#A3648 Sigma-Aldrich). The sections were then treated with target retrieval solution (pH 6.1; S1699, Dako) at 99°C for 40 min and washed with PBS before being used for immunohistochemistry (IHC). The processing of eye globes, isolation and preparation of retinal strips, and retinal immunostaining were extensively detailed in 14,16,19,27 . Briefly, donor eyes were collected within an average of 9 hours after death, puncture at the limbus and fixed in 10% neutral buffered formalin (NBF) or 4% paraformaldehyde (PFA) then stored at 4°C. Fixed eyes were dissected as previously described (same refs as above). Flatmounts were prepared after careful dissection of the eye globes and thorough cleaning of the vitreous humor. Flatmount strips (~2 mm wide) extending diagonally from the optic disc (OD) to the ora serrata (~20–25 mm long) were prepared in 4 predefined regions: Superior Temporal (ST), Inferior Temporal (IT), Inferior Nasal (IN), and Superior Nasal (SN). In this study, we focused our analysis on the ST retinal strip due to the high presence of AD pathology in this region. The flatmount-derived strips were then paraffinized using standard techniques and embedded in paraffin after flip-rotating 90° horizontally. The retinal strips were sectioned (7-10 µm thick) and mounted on microscope slides coated with 3-aminopropyltriethoxysilane. Immunohistochemical analysis Mouse or human brain sections mounted on slides were treated with serum-free protein blocking solution (X0909, Dako), then incubated overnight at 4°C with the primary antibodies (sources and dilutions in Extended Data Table 14). Sections were subsequently incubated for 1 h at room temperature with host-specific fluorophore-conjugated secondary antibodies and coverslipped using ProLong Gold Antifade Mountant with DAPI (Molecular Probes, Life Technologies). In some cases, slides were dipped in Thio-S solution for 1 min (to stain mature Aβ plaques) after the secondary antibody step and then washed in three baths of 70% ethanol (1 min each) before mounting with ProLong Gold DAPI. Negative controls were processed using the same protocol but without a primary antibody to assess nonspecific labelling. Retinal sections were deparaffinized using 100% xylene twice (10 minutes each), rehydrated with decreasing concentrations of ethanol (100% to 70%), and washed with distilled water followed by PBS. After deparaffinization, tissue sections were treated with target antigen retrieval solution (pH 6.1; S1699, Dako) at 98°C for 1 hour and then washed with PBS. Following steps included blocking, primary and secondary antibodies incubations as described above. Golgi-Cox staining Mouse brain hemispheres were processed using a Hito Golgi-Cox OptimStainTM Kit (#HTKNS1125, Hito) according to the manufacturer’s instructions. Briefly, brain samples were immersed in impregnation solution in the dark at room temperature for 2 weeks and then cryoprotected at 4°C.Next, the tissues were embedded in OCT compound, sectioned at a thickness of 100−200 μm, mounted on gelatine-coated slides (#HTHS0102, Hito) and allowed to dry overnight in a dark room.The sections were then incubated in a 20% ammonia solution, dehydrated in ascending concentrations of ethanol (50%, 75%, 95%, and 100%), cleared in xylene, and mounted with Permount mounting medium (#SP15-500, Fisher Scientific). Microscopy and quantification Immunofluorescence images of human and mouse tissue sections were repeatedly captured at 20 × (resolution of 1388 × 1040 pixels, 447.63 µm × 335.30 µm /per image) or 40 × (resolution of 1388 × 1040 pixels, 223.82 µm × 167.70 µm /per image) at the same focal planes with the same exposure time for each marker. Images were randomly acquired as follows: 3 from the central, 4 from the mid-, and 3 from the far-retinal subregions per human retinal strip (one strip per donor); 5–7 images per human brain section (one section per donor); and 5–12 images per mouse brain section (2–3 sections per animal). Images were exported to Fiji ImageJ (version 2.14.0) to analyse parameters of interest. Acquired images were converted to grayscale and standardized to baseline by using a histogram-based threshold in Fiji ImageJ. This baseline-derived threshold was then applied uniformly to the corresponding single channel for all subjects across diagnostic groups. Images were subsequently subjected to ImageJ2/Fiji particle analysis for each biomarker to determine total and % immunoreactive area. To assess levels of H3K9me3 in neuronal versus nonneuronal cells, the intensity of H3K9me3 IR in the nuclei (manually delimited with the polygon tool of ImageJ) of NeuN + (average of 30 nuclei/animal), GFAP + (average of 10 nuclei/animal), or IBA1 + CD45 + (average of 20 nuclei/animal) cells was measured at high magnification. H3K9me3 IR in specific regions of the hippocampus (the cornu ammonis [CA] and dentate gyrus [DG] regions) was quantified within areas manually delimited with the polygon tool of ImageJ or Fiji. In Golgi-Cox–stained brain sections (Fig. 3b,c and Extended Data Fig. 6a), dendrite segments of pyramidal neurons in the cerebral cortex and the CA1 region of the hippocampus were imaged using a Zeiss ApoTome microscope set at 63× (resolution of 1388 × 1040 pixels, 225.56 µm × 169.01 µm /per image). Dendritic spines were classified as thin (filopodia-like and long-thin), mushroom, or stubby types based on established morphological criteria 130 (Fig. 3d), and manually counted. Image capture and quantification analysis were performed by different investigators. Western Blot Analysis Snap-frozen mouse brain tissues were homogenized in radioimmunoprecipitation assay (RIPA) buffer (0.5M Tris-HCl, pH 7.4, 1.5M NaCl, 2.5% deoxycholic acid, 10% NP-40, 10mM EDTA, Millipore; 20–188) supplemented with protease and phosphatase inhibitors. Protein concentration was determined using a bicinchoninic acid protein assay kit (Pierce TM ). Lysates were cleared with brief centrifugation for 10 min at 8000g, normalised, and boiled at 95°C after addition of 6X SDS loading dye. Equal amounts of protein (30 µg per sample) were separated on 4–20% precast polyacrylamide gels (Bio-Rad, catalog #4561094) and transferred to polyvinylidene difluoride membranes. The membranes were then blocked with 2.5% bovine serum albumin in 1x TBS-T (Tris-buffered saline with 0.1% Tween-20) for 1 h at RT, followed by overnight incubation at 4 °C with the primary antibody (sources and dilutions in Extended Data Table 14). After washing with 1x TBS-T, the membrane was incubated with species-specific horseradish peroxidase (HRP)−conjugated or fluorescently labelled secondary antibodies (1:10000). For HRP, the proteins were visualized by incubation with a chemiluminescence substrate kit (#34580, Thermo Fisher). Bands were detected using the LI-COR Odyssey imaging system and quantified using Image Studio software (LI-COR). Relative protein expression levels were calculated by normalizing target protein signals to β-actin or Gapdh. In some cases, membranes were re-probed with primary antibodies after stripping with 1X stripping buffer (NewBlot TM Nitro stripping buffer, Licorbio #928-40030). Complete Blood Count and Hemopexin, Haptoglobin and C3 complement levels Blood was collected from the vena cava while the animals were in a state of deep anaesthesia, just prior to transcardiac perfusion, and transferred to EDTA tubes (BD microtainer #365974). Blood was then processed and analysed with the aid of a hematology analyser according to manufacturer instructions (Horiba ABX Micros 60®). Plasma levels of hemopexin, haptoglobin and C3 complement were assessed using ELISA kits # ab157716, ab157714 and ab157711, respectively, according to the manufacturer’s instructions. Mass spectrometry Mass spectrometry analysis of human and mouse retina and brain tissues, previously reported in 16,55 , followed these steps: (1) preparation of retinal and brain homogenates; (2) tandem mass tag (TMT) labelling; (3) nanoflow liquid chromatography electrospray ionization tandem mass spectrometry; and (4) database searching, peptide quantification, and statistical analysis. Given the exploratory nature and limited sample size of our human and mouse datasets, the following criteria were applied to identify significantly differentially regulated proteins (DEPs) for downstream interpretation: 1) at least three measurements per group; 2) an unadjusted P 1.2. Functional network and computational analysis Heatmaps of detectable protein hierarchies were created, and principal component analysis (PCA) were performed using ClustVis (https://biit.cs.ut.ee/clustvis/) 131 . Volcano plots were prepared using GraphPad Prism 10.6.1. software. Pie charts of protein classifications were created using PANTHER (http://pantherdb.org). Gene Ontology (GO, including Biological Process, Molecular Function and Cellular Component) enrichment analysis were performed using MouseMine (www.mousemine.org) and Metascape (https://metascape.org) databases. Pathway networks created in Metascape were subsequently loaded and modified in Cytoscape 3.10.2 (https://cytoscape.org). DEPs were incorporated into molecular pathway and upstream regulator analyses using Ingenuity Pathway Analysis (IPA, Qiagen; https://digitalinsights.qiagen.com). Enrichment and pathway analysis results are reported with z-scores and Benjamini-Hochberg adjusted p-values to control the FDR. Protein interaction networks were generated in String v12.0 (https://string-db.org) and modified in Cytoscape. Volcano plots representing expression changes [log 2 (FC)] and significance level [-log 10 ( p )] were created using GraphPad Prism v10.6.1. Statistics and reproducibility Statistical analyses were performed using GraphPad Prism v10.6.1. Data was analysed using one- or two-way analysis of variance (ANOVA), followed by Fisher’s least significant difference (LSD) post hoc test for multiple comparisons (mouse data comparisons limited to: DMSO-AD + versus DMSO-WT, ETP69-AD + versus DMSO-AD + and ETP69-WT versus DMSO-WT). Comparisons between two groups were performed using unpaired, two-tailed Student’s t-tests. Correlations were assessed using Pearson’s correlation analysis; Pearson’s coefficient (r) indicates the strength and direction of linear associations. Data is presented as mean ± standard error of the mean (SEM) unless otherwise stated. Violin plots display the median and the lower and upper quartiles. Statistical significance was defined as *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. No statistical methods were used to predetermine sample size. Sample sizes were selected based on prior experience and consistency with previously published studies in the field. No randomization was performed. Data points identified as outliers (2 standard deviations away from the mean) were excluded from analysis prior to statistical testing. Data distribution was assumed to be approximately normal but was not formally tested. Histological and western blot analyses were performed once per experiment and included three or more independent biological replicates. Different investigators performed experiments and analyses at different stages of the study. Investigators were blinded to mouse genotype and treatment during behavioral assays. No randomization was performed. Human and animal experimental groups were balanced for age and sex where possible. Declarations Data availability Most data generated and analysed in this study are included in this manuscript and extended data. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifiers PXD040225 (human data) and PXD041527 (mouse data). Acknowledgements This work was supported by the National Institutes of Health (NIH)/National Institute on Aging (NIA) grants R01AG056478, R01AG055865, and AG056478-04S1 (M.K.H.). The work was also supported by the Hertz Innovation Fund, The Saban, Snyder, Wilstein, and The Gordon Private Foundations, and The Jona Goldrich Center Alzheimer’s Disease (M.K.H.). The Ray Charles Scholar Foundation supported M.R.D. and J.W.W. The authors thank the Cedars-Sinai Biobehavioral Research Core for assistance with and access to equipment for testing. We thank Drs. Carol Ann Miller and Debra Hawes (ADRC Neuropathology Core; University of Southern California), as well as Drs. Rodrigo Medeiros and Joao A. Paulo (University of California-Irvine and Harvard Medical School, respectively) for providing human brain tissues, analysis, and neuropathological reports. We thank Dr. Min Lin (Horne’s lab) for providing ETP69 compound and Dr. Rakez Kayed for previously sharing the T22 oligo-tau antibody. We also thank Samuel Fuchs, Ella Maru Studio, and Biorender.com for illustrations and figure artwork. Author contributions Study conception and design: MKH, DTF, and KLB Live animal experiments: DTF, JPV, JS, YK, OC, and TS Human and mouse tissue collection, isolation, and processing: DTF, AR, JPV, YK, BPG, JS, OC, HS, SS, and MKH Data acquisition, curation, and analysis: DTF, JPV, AR, YK, JS, BPG, SS, HS, MRD, JWW, LSS, and MKH Mass spectrometry experiments and analysis: MM, JPV, JS, DTF, YK, SLG, VKG, and MKH Statistical analysis: JPV, DTF, AR, and MKH Interpretation of the data: JPV, DTF, AR, YK, JS, MF, KLB, and MKH External resources: DH, MTK, KLB Writing-original draft: JPV, DTF and MKH Writing-editing, review: DTF, JPV, AR, YK, JS, BPG, SS, HS, OC, MRD, JWW, SLG, VKG, LS, MTK, TS, MF, DH, MM, KLB, and MKH Study supervision: MKH All authors read and approved the final manuscript. Competing interests KLB is the Co-Chairman and shareholder, and MKH is a scientific advisor, of Fortem Neurosciences, Inc. Unrelated to this study: YK, KLB and MKH are co-founders of NeuroVision Imaging, Inc. The other authors have no conflicts to disclose. References Fraga, M. F. & Esteller, M. Epigenetics and aging: the targets and the marks. Trends Genet 23 , 413-418 (2007). https://doi.org:10.1016/j.tig.2007.05.008 Berson, A., Nativio, R., Berger, S. L. & Bonini, N. M. Epigenetic Regulation in Neurodegenerative Diseases. Trends Neurosci 41 , 587-598 (2018). https://doi.org:10.1016/j.tins.2018.05.005 Delgado-Morales, R., Agis-Balboa, R. C., Esteller, M. & Berdasco, M. Epigenetic mechanisms during ageing and neurogenesis as novel therapeutic avenues in human brain disorders. Clin Epigenetics 9 , 67 (2017). https://doi.org:10.1186/s13148-017-0365-z Nikolac Perkovic, M. et al. Epigenetics of Alzheimer's Disease. Biomolecules 11 (2021). https://doi.org:10.3390/biom11020195 Sen, P., Shah, P. P., Nativio, R. & Berger, S. L. Epigenetic Mechanisms of Longevity and Aging. Cell 166 , 822-839 (2016). https://doi.org:10.1016/j.cell.2016.07.050 Wang, K. et al. Epigenetic regulation of aging: implications for interventions of aging and diseases. Signal Transduct Target Ther 7 , 374 (2022). https://doi.org:10.1038/s41392-022-01211-8 Jeremic, D., Jimenez-Diaz, L. & Navarro-Lopez, J. D. Targeting epigenetics: A novel promise for Alzheimer's disease treatment. Ageing Res Rev 90 , 102003 (2023). https://doi.org:10.1016/j.arr.2023.102003 Nativio, R. et al. An integrated multi-omics approach identifies epigenetic alterations associated with Alzheimer's disease. Nat Genet 52 , 1024-1035 (2020). https://doi.org:10.1038/s41588-020-0696-0 Alzheimer's disease facts and figures. Alzheimers Dement 20 , 3708-3821 (2024). https://doi.org:10.1002/alz.13809 Jack, C. R., Jr. et al. NIA-AA Research Framework: Toward a biological definition of Alzheimer's disease. Alzheimers Dement 14 , 535-562 (2018). https://doi.org:10.1016/j.jalz.2018.02.018 Heneka, M. T. et al. Neuroinflammation in Alzheimer's disease. Lancet Neurol 14 , 388-405 (2015). https://doi.org:10.1016/S1474-4422(15)70016-5 Palop, J. J. & Mucke, L. Amyloid-beta-induced neuronal dysfunction in Alzheimer's disease: from synapses toward neural networks. Nat Neurosci 13 , 812-818 (2010). https://doi.org:10.1038/nn.2583 Koronyo-Hamaoui, M. et al. Identification of amyloid plaques in retinas from Alzheimer's patients and noninvasive in vivo optical imaging of retinal plaques in a mouse model. Neuroimage 54 Suppl 1 , S204-217 (2011). https://doi.org:10.1016/j.neuroimage.2010.06.020 Koronyo, Y. et al. Retinal amyloid pathology and proof-of-concept imaging trial in Alzheimer's disease. JCI Insight 2 (2017). https://doi.org:10.1172/jci.insight.93621 La Morgia, C. et al. Melanopsin retinal ganglion cell loss in Alzheimer disease. Ann Neurol 79 , 90-109 (2016). https://doi.org:10.1002/ana.24548 Koronyo, Y. et al. Retinal pathological features and proteome signatures of Alzheimer's disease. Acta Neuropathol 145 , 409-438 (2023). https://doi.org:10.1007/s00401-023-02548-2 Shi, H. et al. Identification of early pericyte loss and vascular amyloidosis in Alzheimer's disease retina. Acta Neuropathol 139 , 813-836 (2020). https://doi.org:10.1007/s00401-020-02134-w Shi, H. et al. Retinal arterial Abeta(40) deposition is linked with tight junction loss and cerebral amyloid angiopathy in MCI and AD patients. Alzheimers Dement 19 , 5185-5197 (2023). https://doi.org:10.1002/alz.13086 Shi, H. et al. Identification of retinal oligomeric, citrullinated, and other tau isoforms in early and advanced AD and relations to disease status. Acta Neuropathol 148 , 3 (2024). https://doi.org:10.1007/s00401-024-02760-8 Gaire, B. P. et al. Alzheimer's disease pathophysiology in the Retina. Prog Retin Eye Res 101 , 101273 (2024). https://doi.org:10.1016/j.preteyeres.2024.101273 Grimaldi, A. et al. Neuroinflammatory Processes, A1 Astrocyte Activation and Protein Aggregation in the Retina of Alzheimer's Disease Patients, Possible Biomarkers for Early Diagnosis. Front Neurosci 13 , 925 (2019). https://doi.org:10.3389/fnins.2019.00925 Xu, Q. A. et al. Muller cell degeneration and microglial dysfunction in the Alzheimer's retina. Acta Neuropathol Commun 10 , 145 (2022). https://doi.org:10.1186/s40478-022-01448-y Wijesinghe, P. et al. Decoding amyloid beta clearance systems at inner blood-retina barrier using three-dimensional ex vivo retinal imaging in Alzheimer's disease. Alzheimers Dement 21 , e70592 (2025). https://doi.org:10.1002/alz.70592 Walkiewicz, G. et al. Primary retinal tauopathy: A tauopathy with a distinct molecular pattern. Alzheimers Dement 20 , 330-340 (2024). https://doi.org:10.1002/alz.13424 Hart de Ruyter, F. J. et al. Phosphorylated tau in the retina correlates with tau pathology in the brain in Alzheimer's disease and primary tauopathies. Acta Neuropathol 145 , 197-218 (2023). https://doi.org:10.1007/s00401-022-02525-1 Hinton, D. R., Sadun, A. A., Blanks, J. C. & Miller, C. A. Optic-nerve degeneration in Alzheimer's disease. N Engl J Med 315 , 485-487 (1986). https://doi.org:10.1056/NEJM198608213150804 Gaire, B. P. et al. Identification of Chlamydia pneumoniae and NLRP3 inflammasome activation in Alzheimer's disease retina. Nat Commun 17 , 771 (2026). https://doi.org:10.1038/s41467-026-68580-4 Pennington, K. L. & DeAngelis, M. M. Epigenetic Mechanisms of the Aging Human Retina. J Exp Neurosci 9 , 51-79 (2015). https://doi.org:10.4137/JEN.S25513 Xu, C., Fu, X., Qin, H. & Yao, K. Traversing the epigenetic landscape: DNA methylation from retina to brain in development and disease. Front Cell Neurosci 18 , 1499719 (2024). https://doi.org:10.3389/fncel.2024.1499719 Advani, J. et al. QTL mapping of human retina DNA methylation identifies 87 gene-epigenome interactions in age-related macular degeneration. Nat Commun 15 , 1972 (2024). https://doi.org:10.1038/s41467-024-46063-8 Mondal, A. K., Gaur, M., Advani, J. & Swaroop, A. Epigenome-metabolism nexus in the retina: implications for aging and disease. Trends Genet 40 , 718-729 (2024). https://doi.org:10.1016/j.tig.2024.04.012 Bowman, G. D. & Poirier, M. G. Post-translational modifications of histones that influence nucleosome dynamics. Chem Rev 115 , 2274-2295 (2015). https://doi.org:10.1021/cr500350x Kouzarides, T. Chromatin modifications and their function. Cell 128 , 693-705 (2007). https://doi.org:10.1016/j.cell.2007.02.005 Sultan, F. A. & Day, J. J. Epigenetic mechanisms in memory and synaptic function. Epigenomics 3 , 157-181 (2011). https://doi.org:10.2217/epi.11.6 Herre, M. & Korb, E. The chromatin landscape of neuronal plasticity. Curr Opin Neurobiol 59 , 79-86 (2019). https://doi.org:10.1016/j.conb.2019.04.006 Singh, P., Srivas, S. & Thakur, M. K. Epigenetic Regulation of Memory-Therapeutic Potential for Disorders. Curr Neuropharmacol 15 , 1208-1221 (2017). https://doi.org:10.2174/1570159X15666170404144522 Campbell, R. R. & Wood, M. A. How the epigenome integrates information and reshapes the synapse. Nat Rev Neurosci 20 , 133-147 (2019). https://doi.org:10.1038/s41583-019-0121-9 Walker, M. P., LaFerla, F. M., Oddo, S. S. & Brewer, G. J. Reversible epigenetic histone modifications and Bdnf expression in neurons with aging and from a mouse model of Alzheimer's disease. Age (Dordr) 35 , 519-531 (2013). https://doi.org:10.1007/s11357-011-9375-5 Lee, M. Y. et al. Epigenome signatures landscaped by histone H3K9me3 are associated with the synaptic dysfunction in Alzheimer's disease. Aging Cell 19 , e13153 (2020). https://doi.org:10.1111/acel.13153 Zheng, Y. et al. Inhibition of EHMT1/2 rescues synaptic and cognitive functions for Alzheimer's disease. Brain 142 , 787-807 (2019). https://doi.org:10.1093/brain/awy354 Coppede, F. The potential of epigenetic therapies in neurodegenerative diseases. Front Genet 5 , 220 (2014). https://doi.org:10.3389/fgene.2014.00220 Adwan, L. & Zawia, N. H. Epigenetics: a novel therapeutic approach for the treatment of Alzheimer's disease. Pharmacol Ther 139 , 41-50 (2013). https://doi.org:10.1016/j.pharmthera.2013.03.010 Baumann, M. et al. Tricyclic Analogues of Epidithiodioxopiperazine Alkaloids with Promising In Vitro and In Vivo Antitumor Activity. Chem Sci 6 , 4451-4457 (2015). https://doi.org:10.1039/C5SC01536G Fyodorov, D. V., Zhou, B. R., Skoultchi, A. I. & Bai, Y. Emerging roles of linker histones in regulating chromatin structure and function. Nat Rev Mol Cell Biol 19 , 192-206 (2018). https://doi.org:10.1038/nrm.2017.94 Tvardovskiy, A., Schwammle, V., Kempf, S. J., Rogowska-Wrzesinska, A. & Jensen, O. N. Accumulation of histone variant H3.3 with age is associated with profound changes in the histone methylation landscape. Nucleic Acids Res 45 , 9272-9289 (2017). https://doi.org:10.1093/nar/gkx696 Maze, I. et al. Critical Role of Histone Turnover in Neuronal Transcription and Plasticity. Neuron 87 , 77-94 (2015). https://doi.org:10.1016/j.neuron.2015.06.014 Gallegos, D. A., Chan, U., Chen, L. F. & West, A. E. Chromatin Regulation of Neuronal Maturation and Plasticity. Trends Neurosci 41 , 311-324 (2018). https://doi.org:10.1016/j.tins.2018.02.009 Ohzeki, J. et al. KAT7/HBO1/MYST2 Regulates CENP-A Chromatin Assembly by Antagonizing Suv39h1-Mediated Centromere Inactivation. Dev Cell 37 , 413-427 (2016). https://doi.org:10.1016/j.devcel.2016.05.006 Chang, B., Chen, Y., Zhao, Y. & Bruick, R. K. JMJD6 is a histone arginine demethylase. Science 318 , 444-447 (2007). https://doi.org:10.1126/science.1145801 Chapman, P. F. et al. Impaired synaptic plasticity and learning in aged amyloid precursor protein transgenic mice. Nat Neurosci 2 , 271-276 (1999). https://doi.org:10.1038/6374 Butterfield, D. A. Phosphoproteomics of Alzheimer disease brain: Insights into altered brain protein regulation of critical neuronal functions and their contributions to subsequent cognitive loss. Biochim Biophys Acta Mol Basis Dis 1865 , 2031-2039 (2019). https://doi.org:10.1016/j.bbadis.2018.08.035 Cardozo, P. L. et al. Synaptic Elimination in Neurological Disorders. Curr Neuropharmacol 17 , 1071-1095 (2019). https://doi.org:10.2174/1570159X17666190603170511 Oddo, S. et al. Triple-transgenic model of Alzheimer's disease with plaques and tangles: intracellular Abeta and synaptic dysfunction. Neuron 39 , 409-421 (2003). https://doi.org:10.1016/s0896-6273(03)00434-3 Butovsky, O. et al. Glatiramer acetate fights against Alzheimer's disease by inducing dendritic-like microglia expressing insulin-like growth factor 1. Proc Natl Acad Sci U S A 103 , 11784-11789 (2006). https://doi.org:10.1073/pnas.0604681103 Doustar, J. et al. Parallels between retinal and brain pathology and response to immunotherapy in old, late-stage Alzheimer's disease mouse models. Aging Cell 19 , e13246 (2020). https://doi.org:10.1111/acel.13246 Kasindi, A. et al. Glatiramer Acetate Immunomodulation: Evidence of Neuroprotection and Cognitive Preservation. Cells 11 (2022). https://doi.org:10.3390/cells11091578 Koronyo, Y. et al. Therapeutic effects of glatiramer acetate and grafted CD115(+) monocytes in a mouse model of Alzheimer's disease. Brain 138 , 2399-2422 (2015). https://doi.org:10.1093/brain/awv150 Koronyo-Hamaoui, M. et al. Attenuation of AD-like neuropathology by harnessing peripheral immune cells: local elevation of IL-10 and MMP-9. J Neurochem 111 , 1409-1424 (2009). https://doi.org:10.1111/j.1471-4159.2009.06402.x Bernstein, K. E. et al. Angiotensin-converting enzyme overexpression in myelomonocytes prevents Alzheimer's-like cognitive decline. J Clin Invest 124 , 1000-1012 (2014). https://doi.org:10.1172/JCI66541 Koronyo-Hamaoui, M. et al. Peripherally derived angiotensin converting enzyme-enhanced macrophages alleviate Alzheimer-related disease. Brain 143 , 336-358 (2020). https://doi.org:10.1093/brain/awz364 Li, S. et al. Activated Bone Marrow-Derived Macrophages Eradicate Alzheimer's-Related Abeta(42) Oligomers and Protect Synapses. Front Immunol 11 , 49 (2020). https://doi.org:10.3389/fimmu.2020.00049 Alves, V. C., Figueiro-Silva, J., Ferrer, I. & Carro, E. Epigenetic silencing of OR and TAS2R genes expression in human orbitofrontal cortex at early stages of sporadic Alzheimer's disease. Cell Mol Life Sci 80 , 196 (2023). https://doi.org:10.1007/s00018-023-04845-1 Gil, L. et al. Pathological Nuclear Hallmarks in Dentate Granule Cells of Alzheimer's Patients: A Biphasic Regulation of Neurogenesis. Int J Mol Sci 23 (2022). https://doi.org:10.3390/ijms232112873 Davis, M. R. et al. Retinal ganglion cell vulnerability to pathogenic tau in Alzheimer's disease. Acta Neuropathol Commun 13 , 31 (2025). https://doi.org:10.1186/s40478-025-01935-y Schultz, N., Byman, E., Netherlands Brain, B. & Wennstrom, M. Levels of Retinal Amyloid-beta Correlate with Levels of Retinal IAPP and Hippocampal Amyloid-beta in Neuropathologically Evaluated Individuals. J Alzheimers Dis 73 , 1201-1209 (2020). https://doi.org:10.3233/JAD-190868 Santiago, J. et al. Retinal tau phosphorylation in Alzheimer's disease: A mass spectrometry study. Neurobiol Dis 215 , 107057 (2025). https://doi.org:10.1016/j.nbd.2025.107057 Dumitrascu, O. M. et al. Retinal peri-arteriolar versus peri-venular amyloidosis, hippocampal atrophy, and cognitive impairment: exploratory trial. Acta Neuropathol Commun 12 , 109 (2024). https://doi.org:10.1186/s40478-024-01810-2 Dumitrascu, O. M. et al. Sectoral segmentation of retinal amyloid imaging in subjects with cognitive decline. Alzheimers Dement (Amst) 12 , e12109 (2020). https://doi.org:10.1002/dad2.12109 Snigdha, S. et al. H3K9me3 Inhibition Improves Memory, Promotes Spine Formation, and Increases BDNF Levels in the Aged Hippocampus. J Neurosci 36 , 3611-3622 (2016). https://doi.org:10.1523/JNEUROSCI.2693-15.2016 Bloss, E. B. et al. Evidence for reduced experience-dependent dendritic spine plasticity in the aging prefrontal cortex. J Neurosci 31 , 7831-7839 (2011). https://doi.org:10.1523/JNEUROSCI.0839-11.2011 Mahmmoud, R. R. et al. Spatial and Working Memory Is Linked to Spine Density and Mushroom Spines. PLoS One 10 , e0139739 (2015). https://doi.org:10.1371/journal.pone.0139739 Xu, B. et al. Loss of thin spines and small synapses contributes to defective hippocampal function in aged mice. Neurobiol Aging 71 , 91-104 (2018). https://doi.org:10.1016/j.neurobiolaging.2018.07.010 Karpova, N. N. Role of BDNF epigenetics in activity-dependent neuronal plasticity. Neuropharmacology 76 Pt C , 709-718 (2014). https://doi.org:10.1016/j.neuropharm.2013.04.002 Gupta-Agarwal, S. et al. G9a/GLP histone lysine dimethyltransferase complex activity in the hippocampus and the entorhinal cortex is required for gene activation and silencing during memory consolidation. J Neurosci 32 , 5440-5453 (2012). https://doi.org:10.1523/JNEUROSCI.0147-12.2012 Ionescu-Tucker, A. et al. Exercise Reduces H3K9me3 and Regulates Brain Derived Neurotrophic Factor and GABRA2 in an Age Dependent Manner. Front Aging Neurosci 13 , 798297 (2021). https://doi.org:10.3389/fnagi.2021.798297 Amidfar, M., de Oliveira, J., Kucharska, E., Budni, J. & Kim, Y. K. The role of CREB and BDNF in neurobiology and treatment of Alzheimer's disease. Life Sci 257 , 118020 (2020). https://doi.org:10.1016/j.lfs.2020.118020 Jiao, S. S. et al. Brain-derived neurotrophic factor protects against tau-related neurodegeneration of Alzheimer's disease. Transl Psychiatry 6 , e907 (2016). https://doi.org:10.1038/tp.2016.186 O'Bryant, S. E. et al. Brain-derived neurotrophic factor levels in Alzheimer's disease. J Alzheimers Dis 17 , 337-341 (2009). https://doi.org:10.3233/JAD-2009-1051 Song, J. H., Yu, J. T. & Tan, L. Brain-Derived Neurotrophic Factor in Alzheimer's Disease: Risk, Mechanisms, and Therapy. Mol Neurobiol 52 , 1477-1493 (2015). https://doi.org:10.1007/s12035-014-8958-4 Lin, W. J. et al. VGF and Its C-Terminal Peptide TLQP-62 Regulate Memory Formation in Hippocampus via a BDNF-TrkB-Dependent Mechanism. J Neurosci 35 , 10343-10356 (2015). https://doi.org:10.1523/JNEUROSCI.0584-15.2015 Rouillard, A. D. et al. The harmonizome: a collection of processed datasets gathered to serve and mine knowledge about genes and proteins. Database (Oxford) 2016 (2016). https://doi.org:10.1093/database/baw100 Trani, E. et al. Isolation and characterization of VGF peptides in rat brain. Role of PC1/3 and PC2 in the maturation of VGF precursor. J Neurochem 81 , 565-574 (2002). https://doi.org:10.1046/j.1471-4159.2002.00842.x Alder, J. et al. Brain-derived neurotrophic factor-induced gene expression reveals novel actions of VGF in hippocampal synaptic plasticity. J Neurosci 23 , 10800-10808 (2003). Li, C. et al. Neuropeptide VGF C-Terminal Peptide TLQP-62 Alleviates Lipopolysaccharide-Induced Memory Deficits and Anxiety-like and Depression-like Behaviors in Mice: The Role of BDNF/TrkB Signaling. ACS Chem Neurosci 8 , 2005-2018 (2017). https://doi.org:10.1021/acschemneuro.7b00154 Lin, W. J. et al. An increase in VGF expression through a rapid, transcription-independent, autofeedback mechanism improves cognitive function. Transl Psychiatry 11 , 383 (2021). https://doi.org:10.1038/s41398-021-01489-2 Behnke, J. et al. Neuropeptide VGF Promotes Maturation of Hippocampal Dendrites That Is Reduced by Single Nucleotide Polymorphisms. Int J Mol Sci 18 (2017). https://doi.org:10.3390/ijms18030612 Thakker-Varia, S. et al. VGF (TLQP-62)-induced neurogenesis targets early phase neural progenitor cells in the adult hippocampus and requires glutamate and BDNF signaling. Stem Cell Res 12 , 762-777 (2014). https://doi.org:10.1016/j.scr.2014.03.005 Takei, N. & Nawa, H. mTOR signaling and its roles in normal and abnormal brain development. Front Mol Neurosci 7 , 28 (2014). https://doi.org:10.3389/fnmol.2014.00028 Bartolomucci, A. et al. The extended granin family: structure, function, and biomedical implications. Endocr Rev 32 , 755-797 (2011). https://doi.org:10.1210/er.2010-0027 Tasaki, S., Gaiteri, C., Mostafavi, S., De Jager, P. L. & Bennett, D. A. The Molecular and Neuropathological Consequences of Genetic Risk for Alzheimer's Dementia. Front Neurosci 12 , 699 (2018). https://doi.org:10.3389/fnins.2018.00699 Bai, B. et al. Deep Multilayer Brain Proteomics Identifies Molecular Networks in Alzheimer's Disease Progression. Neuron 105 , 975-991 e977 (2020). https://doi.org:10.1016/j.neuron.2019.12.015 Carrette, O. et al. A panel of cerebrospinal fluid potential biomarkers for the diagnosis of Alzheimer's disease. Proteomics 3 , 1486-1494 (2003). https://doi.org:10.1002/pmic.200300470 Khoonsari, P. E. et al. Improved Differential Diagnosis of Alzheimer's Disease by Integrating ELISA and Mass Spectrometry-Based Cerebrospinal Fluid Biomarkers. J Alzheimers Dis 67 , 639-651 (2019). https://doi.org:10.3233/JAD-180855 Pedrero-Prieto, C. M. et al. A comprehensive systematic review of CSF proteins and peptides that define Alzheimer's disease. Clin Proteomics 17 , 21 (2020). https://doi.org:10.1186/s12014-020-09276-9 Holtta, M. et al. An integrated workflow for multiplex CSF proteomics and peptidomics-identification of candidate cerebrospinal fluid biomarkers of Alzheimer's disease. J Proteome Res 14 , 654-663 (2015). https://doi.org:10.1021/pr501076j Hendrickson, R. C. et al. High Resolution Discovery Proteomics Reveals Candidate Disease Progression Markers of Alzheimer's Disease in Human Cerebrospinal Fluid. PLoS One 10 , e0135365 (2015). https://doi.org:10.1371/journal.pone.0135365 Llano, D. A., Bundela, S., Mudar, R. A., Devanarayan, V. & Alzheimer's Disease Neuroimaging, I. A multivariate predictive modeling approach reveals a novel CSF peptide signature for both Alzheimer's Disease state classification and for predicting future disease progression. PLoS One 12 , e0182098 (2017). https://doi.org:10.1371/journal.pone.0182098 Duits, F. H. et al. Synaptic proteins in CSF as potential novel biomarkers for prognosis in prodromal Alzheimer's disease. Alzheimers Res Ther 10 , 5 (2018). https://doi.org:10.1186/s13195-017-0335-x Sathe, G. et al. Quantitative Proteomic Profiling of Cerebrospinal Fluid to Identify Candidate Biomarkers for Alzheimer's Disease. Proteomics Clin Appl 13 , e1800105 (2019). https://doi.org:10.1002/prca.201800105 Beckmann, N. D. et al. Multiscale causal networks identify VGF as a key regulator of Alzheimer's disease. Nat Commun 11 , 3942 (2020). https://doi.org:10.1038/s41467-020-17405-z Spellman, D. S. et al. Development and evaluation of a multiplexed mass spectrometry based assay for measuring candidate peptide biomarkers in Alzheimer's Disease Neuroimaging Initiative (ADNI) CSF. Proteomics Clin Appl 9 , 715-731 (2015). https://doi.org:10.1002/prca.201400178 Jahn, H. et al. Peptide fingerprinting of Alzheimer's disease in cerebrospinal fluid: identification and prospective evaluation of new synaptic biomarkers. PLoS One 6 , e26540 (2011). https://doi.org:10.1371/journal.pone.0026540 Wang, X. et al. Deciphering cellular transcriptional alterations in Alzheimer's disease brains. Mol Neurodegener 15 , 38 (2020). https://doi.org:10.1186/s13024-020-00392-6 Park, S. A. et al. SWATH-MS analysis of cerebrospinal fluid to generate a robust battery of biomarkers for Alzheimer's disease. Sci Rep 10 , 7423 (2020). https://doi.org:10.1038/s41598-020-64461-y Fargali, S. et al. The granin VGF promotes genesis of secretory vesicles, and regulates circulating catecholamine levels and blood pressure. FASEB J 28 , 2120-2133 (2014). https://doi.org:10.1096/fj.13-239509 Gondre-Lewis, M. C., Park, J. J. & Loh, Y. P. Cellular mechanisms for the biogenesis and transport of synaptic and dense-core vesicles. Int Rev Cell Mol Biol 299 , 27-115 (2012). https://doi.org:10.1016/B978-0-12-394310-1.00002-3 Hariri, A. R. et al. Brain-derived neurotrophic factor val66met polymorphism affects human memory-related hippocampal activity and predicts memory performance. J Neurosci 23 , 6690-6694 (2003). https://doi.org:10.1523/JNEUROSCI.23-17-06690.2003 Egan, M. F. et al. The BDNF val66met polymorphism affects activity-dependent secretion of BDNF and human memory and hippocampal function. Cell 112 , 257-269 (2003). https://doi.org:10.1016/s0092-8674(03)00035-7 Bui, T. M., Wiesolek, H. L. & Sumagin, R. ICAM-1: A master regulator of cellular responses in inflammation, injury resolution, and tumorigenesis. J Leukoc Biol 108 , 787-799 (2020). https://doi.org:10.1002/JLB.2MR0220-549R Maier, M. et al. Complement C3 deficiency leads to accelerated amyloid beta plaque deposition and neurodegeneration and modulation of the microglia/macrophage phenotype in amyloid precursor protein transgenic mice. J Neurosci 28 , 6333-6341 (2008). https://doi.org:10.1523/JNEUROSCI.0829-08.2008 Frenkel, D., Maron, R., Burt, D. S. & Weiner, H. L. Nasal vaccination with a proteosome-based adjuvant and glatiramer acetate clears beta-amyloid in a mouse model of Alzheimer disease. J Clin Invest 115 , 2423-2433 (2005). https://doi.org:10.1172/JCI23241 Bakalash, S. et al. Egr1 expression is induced following glatiramer acetate immunotherapy in rodent models of glaucoma and Alzheimer's disease. Invest Ophthalmol Vis Sci 52 , 9033-9046 (2011). https://doi.org:10.1167/iovs.11-7498 Zuroff, L., Daley, D., Black, K. L. & Koronyo-Hamaoui, M. Clearance of cerebral Abeta in Alzheimer's disease: reassessing the role of microglia and monocytes. Cell Mol Life Sci 74 , 2167-2201 (2017). https://doi.org:10.1007/s00018-017-2463-7 Lebson, L. et al. Trafficking CD11b-positive blood cells deliver therapeutic genes to the brain of amyloid-depositing transgenic mice. J Neurosci 30 , 9651-9658 (2010). https://doi.org:10.1523/JNEUROSCI.0329-10.2010 Theriault, P., ElAli, A. & Rivest, S. The dynamics of monocytes and microglia in Alzheimer's disease. Alzheimers Res Ther 7 , 41 (2015). https://doi.org:10.1186/s13195-015-0125-2 Rosenzweig, N. et al. PD-1/PD-L1 checkpoint blockade harnesses monocyte-derived macrophages to combat cognitive impairment in a tauopathy mouse model. Nat Commun 10 , 465 (2019). https://doi.org:10.1038/s41467-019-08352-5 Munoz-Castro, C. et al. Monocyte-derived cells invade brain parenchyma and amyloid plaques in human Alzheimer's disease hippocampus. Acta Neuropathol Commun 11 , 31 (2023). https://doi.org:10.1186/s40478-023-01530-z Deczkowska, A., Amit, I. & Schwartz, M. Microglial immune checkpoint mechanisms. Nat Neurosci 21 , 779-786 (2018). https://doi.org:10.1038/s41593-018-0145-x Butovsky, O., Kunis, G., Koronyo-Hamaoui, M. & Schwartz, M. Selective ablation of bone marrow-derived dendritic cells increases amyloid plaques in a mouse Alzheimer's disease model. Eur J Neurosci 26 , 413-416 (2007). https://doi.org:10.1111/j.1460-9568.2007.05652.x Rentsendorj, A. et al. A novel role for osteopontin in macrophage-mediated amyloid-beta clearance in Alzheimer's models. Brain Behav Immun 67 , 163-180 (2018). https://doi.org:10.1016/j.bbi.2017.08.019 Helwig, M. et al. The neuroendocrine protein 7B2 suppresses the aggregation of neurodegenerative disease-related proteins. J Biol Chem 288 , 1114-1124 (2013). https://doi.org:10.1074/jbc.M112.417071 Hoshino, A. et al. A novel function for proSAAS as an amyloid anti-aggregant in Alzheimer's disease. J Neurochem 128 , 419-430 (2014). https://doi.org:10.1111/jnc.12454 El Gaamouch, F. et al. VGF-derived peptide TLQP-21 modulates microglial function through C3aR1 signaling pathways and reduces neuropathology in 5xFAD mice. Mol Neurodegener 15 , 4 (2020). https://doi.org:10.1186/s13024-020-0357-x Elmadany, N. et al. The VGF-derived Peptide TLQP21 Impairs Purinergic Control of Chemotaxis and Phagocytosis in Mouse Microglia. J Neurosci 40 , 3320-3331 (2020). https://doi.org:10.1523/JNEUROSCI.1458-19.2020 Cho, K. et al. TLQP-21 mediated activation of microglial BV2 cells promotes clearance of extracellular fibril amyloid-beta. Biochem Biophys Res Commun 524 , 764-771 (2020). https://doi.org:10.1016/j.bbrc.2020.01.111 Lee, J., Han, M., Wang, K., Butler, L. R. & Sinclair, D. A. Epigenetic reprogramming for ocular aging and disease: Mechanisms, biomarkers, and the road to the clinic. Prog Retin Eye Res 111 , 101442 (2026). https://doi.org:10.1016/j.preteyeres.2026.101442 Pelzel, H. R. & Nickells, R. W. A role for epigenetic changes in the development of retinal neurodegenerative conditions. J Ocul Biol Dis Infor 4 , 104-110 (2011). https://doi.org:10.1007/s12177-012-9079-9 Vit, J. P. et al. Color and contrast vision in mouse models of aging and Alzheimer's disease using a novel visual-stimuli four-arm maze. Sci Rep 11 , 1255 (2021). https://doi.org:10.1038/s41598-021-80988-0 Vit, J. P. et al. Visual-stimuli Four-arm Maze test to Assess Cognition and Vision in Mice. Bio Protoc 11 , e4234 (2021). https://doi.org:10.21769/BioProtoc.4234 Risher, W. C., Ustunkaya, T., Singh Alvarado, J. & Eroglu, C. Rapid Golgi analysis method for efficient and unbiased classification of dendritic spines. PLoS One 9 , e107591 (2014). https://doi.org:10.1371/journal.pone.0107591 Metsalu, T. & Vilo, J. ClustVis: a web tool for visualizing clustering of multivariate data using Principal Component Analysis and heatmap. Nucleic Acids Res 43 , W566-570 (2015). https://doi.org:10.1093/nar/gkv468 Tables Table 1. Demographic and brain neuropathology data of human subjects used for cerebral and retinal histopathological analysis. Diagnosis One-Way ANOVA CN MCI AD F-value P-value Subjects 6 4F/2M 7 4F/3M 13 7F/6M N/A N/A Age 91.3 ± 2.8 91.1 ± 1.9 85.5 ± 2.7 1.612 0.2212 Race 2H/4W 33/67% 1B/6W 14/86% 3A+B/10W 23/77% N/A N/A MMSE 28.2 ± 1.1 22.8 ± 2.9 16.3 ± 1.7 9.601 0.0012 CDR 0.17 ± 0.17 1.50 ± 0.44 2.27 ± 0.26 10.94 0.0005 Braak stage Stage I-II Stage III-IV Stage V-VI 2.7 ± 0.7 3 2 1 3.6 ± 0.6 2 2 3 5.2 ± 0.2 0 2 11 8.674 0.0016 ABC score 1.95 ± 0.30 2.06 ± 0.24 2.75 ± 0.09 6.472 0.0059 Aβ score (A9) 1.95 ± 0.41 2.11 ± 0.42 3.82 ± 0.48 5.340 0.0133 NFT score (A9) 0.08 ± 0.08 1.21 ± 0.49 1.82 ± 0.57 2.680 0.0919 Atrophy score (A9) 1.50 ± 0.50 1.57 ± 0.53 2.50 ± 0.50 1.219 0.3148 Where applicable, data are presented as mean ± SEM. CN—cognitively normal, MCI—mild cognitive impairment, AD—Alzheimer’s disease. F—female, M—male. Age at death in years. H—Hispanic, W—White, B—Black, A—Asian. MMSE—mini mental state examination. CDR—clinical dementia rating: 0—normal cognition, 1—mild dementia, 2—moderate dementia, 3—severe dementia. ABC score: A—Aβ plaque score modified from Thal, B—NFT stage modified from Braak, C—Neuritic plaque score modified from CERAD. Aβ, NFT and atrophy severity scores: 0—none, 1—sparse, 3—moderate, 5—frequent. A9—Brodmann area A9, dorsolateral prefrontal cortex. N/A—not applicable. Additional Declarations Yes there is potential Competing Interest. KLB is the Co-Chairman and shareholder, and MKH is a scientific advisor, of Fortem Neurosciences, Inc. Unrelated to this study: YK, KLB and MKH are co-founders of NeuroVision Imaging, Inc. The other authors have no conflicts to disclose. Supplementary Files FuchsVitRentsendorjetalExtendedDataFile3.11.2026F.pdf Extended Data Figures and Tables Extended Data Figures 1 to 14 Extended Data Tables 1 to 14 Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8913130","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":605260763,"identity":"06b148c5-8c2e-42c0-af61-424d1e354488","order_by":0,"name":"Maya Koronyo-Hamaoui","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2klEQVRIiWNgGAWjYNCCA1D6A5SWIFoL4wyStTDzEKNFt/2M4QeGMzb5/NI9ho9tauqi+RuYD97mwaPF7EyOsQTDjTTLmXPOGBvnHDucO+MAW7I1Xi0HcgwkGD4cNjC4kWMmndtwIHcDA4+ZNF4t598Y/4BrsWyoA2rh/4ZfC1Al0GFQLYwNzCBb2AhoeVZmkXAmzUByRlqxYQ/IL4fZjC3n4HVY8uYbH47ZGPBLJG988KOmLre/vfnhjTd4tDAwcBgwJMAYYMCMVzkIsD9AZ4yCUTAKRsEoQAUAn11NQgXvoTAAAAAASUVORK5CYII=","orcid":"","institution":"Cedars Sinai","correspondingAuthor":true,"prefix":"","firstName":"Maya","middleName":"","lastName":"Koronyo-Hamaoui","suffix":""},{"id":605260764,"identity":"0a2df92f-2e0f-478c-a88a-a082fa326748","order_by":1,"name":"Dieu-Trang Fuchs","email":"","orcid":"","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Dieu-Trang","middleName":"","lastName":"Fuchs","suffix":""},{"id":605260765,"identity":"f4444838-ba1b-427b-99d5-b5706b7e9546","order_by":2,"name":"Jean-Philippe Vit","email":"","orcid":"","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Jean-Philippe","middleName":"","lastName":"Vit","suffix":""},{"id":605260766,"identity":"3001cda2-cba5-4016-b9f4-f934f4895bae","order_by":3,"name":"Altan Rentsendorj","email":"","orcid":"","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Altan","middleName":"","lastName":"Rentsendorj","suffix":""},{"id":605260767,"identity":"4bde50ea-9eed-4862-9c0b-e61c9229499d","order_by":4,"name":"Yosef Koronyo","email":"","orcid":"","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Yosef","middleName":"","lastName":"Koronyo","suffix":""},{"id":605260768,"identity":"05bcbe29-cd32-49ee-b7d6-df5144c9378f","order_by":5,"name":"Julia Sheyn","email":"","orcid":"","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Julia","middleName":"","lastName":"Sheyn","suffix":""},{"id":605260769,"identity":"5e10f8a3-f0fd-4a67-861a-571f38b48313","order_by":6,"name":"Bhakta Prasad Gaire","email":"","orcid":"https://orcid.org/0000-0002-9738-6493","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Bhakta","middleName":"Prasad","lastName":"Gaire","suffix":""},{"id":605260770,"identity":"8ed48e9a-7541-44b3-9090-660cccf4adec","order_by":7,"name":"Saba Shahin","email":"","orcid":"https://orcid.org/0000-0002-0576-9383","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Saba","middleName":"","lastName":"Shahin","suffix":""},{"id":605260771,"identity":"a35da268-3287-4525-8c19-c970ba5946a9","order_by":8,"name":"Haoshen Shi","email":"","orcid":"","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Haoshen","middleName":"","lastName":"Shi","suffix":""},{"id":605260772,"identity":"911c44e8-608c-43ab-92a4-1504f2b3db47","order_by":9,"name":"Oksana Chepurna","email":"","orcid":"","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Oksana","middleName":"","lastName":"Chepurna","suffix":""},{"id":605260773,"identity":"2dc54284-9879-47d3-9f8f-e7f9099e44dc","order_by":10,"name":"Miyah Davis","email":"","orcid":"","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Miyah","middleName":"","lastName":"Davis","suffix":""},{"id":605260774,"identity":"b1f7accf-36a9-460f-9539-338d6c0aa44c","order_by":11,"name":"Jered Wilson","email":"","orcid":"","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Jered","middleName":"","lastName":"Wilson","suffix":""},{"id":605260775,"identity":"bd02f522-c4f4-4af9-b0fe-da8061a6eb4f","order_by":12,"name":"Stuart Graham","email":"","orcid":"https://orcid.org/0000-0001-7519-969X","institution":"Macquarie University","correspondingAuthor":false,"prefix":"","firstName":"Stuart","middleName":"","lastName":"Graham","suffix":""},{"id":605260776,"identity":"4a2feeb4-0d24-4fa3-a29b-347b60f4b31f","order_by":13,"name":"Vivek Gupta","email":"","orcid":"","institution":"Macquarie University","correspondingAuthor":false,"prefix":"","firstName":"Vivek","middleName":"","lastName":"Gupta","suffix":""},{"id":605260777,"identity":"8db36d00-6115-490c-863d-f2578c354e4f","order_by":14,"name":"Lon S. Schneider","email":"","orcid":"","institution":"University of Southern California","correspondingAuthor":false,"prefix":"","firstName":"Lon","middleName":"S.","lastName":"Schneider","suffix":""},{"id":605260778,"identity":"60ce743a-6e45-4ec8-98e9-c02054fc721a","order_by":15,"name":"Michael Kleinman","email":"","orcid":"","institution":"University of California Irvine","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"","lastName":"Kleinman","suffix":""},{"id":605260779,"identity":"e888d319-5e78-4b87-a2b4-80498416f732","order_by":16,"name":"Tao Sun","email":"","orcid":"","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Tao","middleName":"","lastName":"Sun","suffix":""},{"id":605260780,"identity":"d583377d-4fd9-4b91-a6bc-edd27245af69","order_by":17,"name":"Margaret Fahnestock","email":"","orcid":"https://orcid.org/0000-0002-6815-0529","institution":"McMaster University","correspondingAuthor":false,"prefix":"","firstName":"Margaret","middleName":"","lastName":"Fahnestock","suffix":""},{"id":605260781,"identity":"21bef49d-7c19-464e-9922-ca53b69e1cb3","order_by":18,"name":"David Horne","email":"","orcid":"https://orcid.org/0000-0002-9300-1082","institution":"Beckman Research Institute of the City of Hope","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Horne","suffix":""},{"id":605260782,"identity":"cfcccadf-037e-4128-965b-d4fa23c145cb","order_by":19,"name":"Mehdi Mirzaei","email":"","orcid":"","institution":"Macquarie University","correspondingAuthor":false,"prefix":"","firstName":"Mehdi","middleName":"","lastName":"Mirzaei","suffix":""},{"id":605260783,"identity":"3b7a0a17-5bff-451e-bf4f-1324c6a50c67","order_by":20,"name":"Keith Black","email":"","orcid":"https://orcid.org/0000-0002-0546-4934","institution":"Cedars-Sinai Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Keith","middleName":"","lastName":"Black","suffix":""}],"badges":[],"createdAt":"2026-02-19 01:45:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8913130/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8913130/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104780857,"identity":"27134c0f-96fa-4fe8-9b07-98f49a2030b4","added_by":"auto","created_at":"2026-03-17 07:54:07","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1119753,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIncreased cerebral and retinal H3K9me3 coincides with cognitive deficits and AD-related neuropathology\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, Immunofluorescence images for H3K9me3 (red), 4G8-Aβ plaques (green), and DAPI nuclei (blue) in the dorsolateral prefrontal cortex (Brodmann area 9; A9) of patients with AD or mild cognitive impairment (MCI), and cognitively normal subjects (CN). Scale bar, 100 μm. \u003cstrong\u003eb\u003c/strong\u003e, Quantification of H3K9me3 immunoreactivity (IR) corrected for DAPI nuclei counts (area, μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e2\u003c/sup\u003e/DAPI) in 11 AD and 5 MCI compared to 6 CN patients. \u003cstrong\u003ec–f\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003ePearson’s correlations of A9 H3K9me3/DAPI with clinical dementia rating (CDR, n = 22) (\u003cstrong\u003ec\u003c/strong\u003e), mini mental state examination (MMSE, n = 19) (\u003cstrong\u003ed\u003c/strong\u003e), A9 NFT scores (n = 21) (\u003cstrong\u003ee\u003c/strong\u003e), and A9 Aβ scores (n = 21) (\u003cstrong\u003ef\u003c/strong\u003e). Correlations limited to CN-MCI individuals are shown in blue for CDR, MMSE and NFT scores. \u003cstrong\u003eg\u003c/strong\u003e, Immunofluorescence images for H3K9me3 (red), 4G8-Aβ plaques (green), and DAPI nuclei (blue) in the retina of AD, MCI and CN subjects. Scale bar, 50 μm. \u003cstrong\u003eh\u003c/strong\u003e, Quantification of retinal H3K9me3/DAPI (area, μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e/DAPI) in 12 AD, 6 MCI and 6 CN donors. \u003cstrong\u003ei–k\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003ePearson’s correlations between retinal H3K9me3/DAPI and retinal 4G8-Aβ\u0026nbsp; (% area, n = 18) or T22 oligo-tau (% area, n = 17) (\u003cstrong\u003ei\u003c/strong\u003e), between retinal and cerebral H3K9me3/DAPI (n = 21) (\u003cstrong\u003ej\u003c/strong\u003e), and between retinal H3K9me3/DAPI and CDR (n = 23) or MMSE (n = 19) (\u003cstrong\u003ek\u003c/strong\u003e). \u003cstrong\u003el\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eGene Ontology analysis of DEPs in the temporal cortex of AD patients pertaining to chromatin organization and epigenetic regulation of gene expression. \u003cstrong\u003em\u003c/strong\u003e, Expression profiles of select first and second neighbours upstream and downstream of SUV39H1 predicting its activation in the brain of AD patients, according to IPA. \u003cstrong\u003en\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eGene Ontology analysis of upregulated DEPs in the retina of AD patients related to heterochromatin formation and epigenetic regulation of gene expression. \u003cstrong\u003eo\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eNormalised expression profiles of DEPs in the retina of AD patients versus CN individuals, related to histone H3 modifying processes, including methyltransferase, demethylase and acetyltransferase activities. Coloured diamonds indicate specificity for epigenetic histone modification marks. Individual data points are presented with group means ± SEMs. # p \u0026lt; 0.05, ## p \u0026lt; 0.01 and #### p \u0026lt; 0.0001, by one-way ANOVA followed by\u0026nbsp;Fisher’s LSD \u003cem\u003epost hoc\u003c/em\u003e\u0026nbsp;test.\u003c/p\u003e","description":"","filename":"Fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-8913130/v1/24d46b0c8e4777e392c7e401.png"},{"id":104538709,"identity":"ea540d94-f53c-4c51-bfad-fb3c191a5c6b","added_by":"auto","created_at":"2026-03-13 04:58:06","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1210033,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eShort- and long-term effects of ETP69 on cognitive functions and AD-related pathology in aged mice\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, Molecular structure of ETP69. \u003cstrong\u003eb\u003c/strong\u003e, Behavioural timeline and ETP69 (10 mg/kg) or vehicle administration in 18-month-old AD\u003csup\u003e+\u003c/sup\u003e and WT mice under single (S: 21 DMSO-WT, 22 ETP69-WT, 15 DMSO-AD\u003csup\u003e+\u003c/sup\u003e,\u003csup\u003e \u003c/sup\u003e18 ETP69-AD\u003csup\u003e+\u003c/sup\u003e), repeated (R: 7 DMSO-WT, 6 ETP69-WT, 7 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 6 ETP69-AD\u003csup\u003e+\u003c/sup\u003e) or boost (B: 9 DMSO-WT, 10 ETP69-WT, 7 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 8 ETP69-AD\u003csup\u003e+\u003c/sup\u003e) regimens. \u003cstrong\u003ec,d\u003c/strong\u003e, Single regimen. Total entries (\u003cstrong\u003ec\u003c/strong\u003e) and percentage of alternations (\u003cstrong\u003ed\u003c/strong\u003e) in Y-maze and colour/contrast X-maze. \u003cstrong\u003ee\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eS and R regimens. percentages of entries into blue arm and blue«white bidirectional transitions in colour X-maze. \u003cstrong\u003ef–h\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eCombined S, R and B regimens. \u003cstrong\u003ef\u003c/strong\u003e, Number of errors during acquisition, memory retention and reversal phases in the Barnes maze. \u003cstrong\u003eg,\u003c/strong\u003e Chord diagrams of mean trajectories during retention and reversal phases. \u003cstrong\u003eh\u003c/strong\u003e, Freezing time in the contextual fear conditioning. \u003cstrong\u003ei\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eImmunofluorescence images of H3K9me3 (red), GFAP-astrocytes (green) and 6E10-Aβ plaques (blue) IR with quantifications of H3K9me3 IR (μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e) in the cerebral cortex (CC). Scale bar, 100 μm. \u003cstrong\u003ej\u003c/strong\u003e, Western blot image and quantification of H3K9me3/β-actin in the hippocampus. \u003cstrong\u003ek\u003c/strong\u003e, Immunofluorescence images H3K9me3 (red), GFAP (green) and 6E10-Aβ (blue) with quantification of H3K9me3 IR (μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e) in the hippocampus. Scale bar, 100 μm. \u003cstrong\u003el\u003c/strong\u003e, Quantifications of GFAP and 6E10-Aβ IR (μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e) in the cerebral cortex. \u003cstrong\u003em\u003c/strong\u003e, Western blot image and quantification of GFAP/β-actin in the hippocampus. \u003cstrong\u003en\u003c/strong\u003e, Quantifications of GFAP and 6E10-Aβ IR (μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e) in the hippocampus. \u003cstrong\u003eo\u003c/strong\u003e, Pearson’s correlations of cortical H3K9me3 IR with performance in colour X-maze and with 6E10-Aβ IR, and between hippocampal H3K9me3 IR and performances in Barnes maze and contextual fear conditioning.\u003cstrong\u003e \u003c/strong\u003eD, DMSO; E, ETP69. Black circled numbers represent end-point days. Individual data points are presented with group means ± SEMs. Filled and empty circles represent male and female mice, respectively. Violin plots display median and lower and upper quartiles. # p \u0026lt; 0.05, ## p \u0026lt; 0.01, ###\u003cem\u003e \u003c/em\u003ep \u0026lt; 0.001, and #### p \u0026lt; 0.0001: DMSO-AD\u003csup\u003e+\u003c/sup\u003e versus DMSO-WT mice; * p \u0026lt; 0.05, ** p\u003cem\u003e \u003c/em\u003e\u0026lt; 0.01, *** p \u0026lt; 0.001, and **** p \u0026lt; 0.0001: ETP69 versus DMSO mice, by one- or two-way ANOVA followed by\u0026nbsp;Fisher’s LSD \u003cem\u003epost hoc\u003c/em\u003e\u0026nbsp;test.\u003c/p\u003e","description":"","filename":"Fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-8913130/v1/11856c948c0b65fbb68572ca.png"},{"id":104780782,"identity":"95f50d75-7685-4401-bc06-7e4fbbba21db","added_by":"auto","created_at":"2026-03-17 07:53:57","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1099319,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDendritic spine formation and synaptic integrity in the cortex and hippocampus of aged WT and AD\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e+\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e mice in response to ETP69.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, Simplified experimental timeline in 18-month-old mice (4 DMSO-WT, 5 ETP69-WT, 3 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 4 ETP69-AD\u003csup\u003e+\u003c/sup\u003e). \u003cstrong\u003eb\u003c/strong\u003e, Low- (scale bar, 200 μm), medium- (scale bar, 100 μm), and high-magnification (scale bar, 20 μm) images of Golgi-Cox staining of neurons in the cerebral cortex (CC). \u003cstrong\u003ec\u003c/strong\u003e, Low- (scale bar, 200 μm) and high-magnification (scale bar, 20 μm) images of Golgi-Cox staining in the hippocampus (Hipp). \u003cstrong\u003ed\u003c/strong\u003e, Illustration and high-magnification (scale bar, 5 μm) photograph of dendritic spines classified as thin (filopodia-like and long-thin; T), stubby (S), and mushroom (M) spines according to size and shape. \u003cstrong\u003ee\u003c/strong\u003e, Density of thin and stubby spines as counts per 100 μm of dendrite in the cerebral cortex and hippocampus. \u003cstrong\u003ef\u003c/strong\u003e, Density of dendritic spines as counts per 100 μm of dendrite. \u003cstrong\u003eg\u003c/strong\u003e, Pearson’s correlations between cortical or hippocampal thin spine density and behavioural performance in the Barnes maze or in the contextual fear conditioning. \u003cstrong\u003eh\u003c/strong\u003e, Ratio of thin spines to total spines. \u003cstrong\u003ei\u003c/strong\u003e, Pearson’s correlation between the cortical thin spine ratio and animal’s performance in the retention phase of the Barnes maze test. \u003cstrong\u003ej\u003c/strong\u003e, Immunofluorescence images and quantification of post-synaptic marker PSD95 IR (red, μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e) in the hippocampus (5 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 6 ETP69-AD\u003csup\u003e+\u003c/sup\u003e). Scale bar, 200 μm. DAPI nuclei (blue). \u003cstrong\u003ek\u003c/strong\u003e, Western blot image and normalised quantification of presynaptic marker synaptophysin (SYP)/β-actin expression in the hippocampus (n = 3 per group). D, DMSO; E, ETP69. Individual data points are presented with group means ± SEMs. Median, lower and upper quartiles are indicated on violin plots. ## p \u0026lt; 0.01 and ### p \u0026lt; 0.001: DMSO-AD\u003csup\u003e+\u003c/sup\u003e versus DMSO-WT mice; \u003cstrong\u003e*\u003c/strong\u003e p \u0026lt; 0.05, \u003cstrong\u003e**\u003c/strong\u003e p\u003cem\u003e \u003c/em\u003e\u0026lt; 0.01, *** p \u0026lt; 0.001, and \u003cstrong\u003e****\u003c/strong\u003e p \u0026lt; 0.0001: ETP69 versus DMSO mice; by one-way ANOVA followed by\u003cem\u003e \u003c/em\u003eFisher’s LSD \u003cem\u003epost hoc\u003c/em\u003e\u0026nbsp;test or two-tailed unpaired Student’s t-test.\u003c/p\u003e","description":"","filename":"Fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-8913130/v1/09c0924cd79efb2dba090fb5.png"},{"id":104538705,"identity":"85c1b236-514f-48a3-8006-34118cb7ed51","added_by":"auto","created_at":"2026-03-13 04:58:06","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1139501,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eImpact of ETP69 on AD-like pathology and immune responses in 14-month-old triple or double transgenic AD\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e+\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e mice.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, Timeline for a single i.p. injection of ETP69 or vehicle in 3xTg-AD mice (5 DMSO-3xTg-AD, 6 ETP69-3xTg-AD). \u003cstrong\u003eb\u003c/strong\u003e, Immunofluorescence images and quantifications of H3K9me3 (red, μm\u003csup\u003e2\u003c/sup\u003e×100) and PHF-tau (white, μm\u003csup\u003e2\u003c/sup\u003e) IR in the cerebral cortex of 3xTg-AD mice. Scale bar, 20 μm. \u003cstrong\u003ec\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003ePearson’s correlation between cortical H3K9me3 and PHF-tau levels. \u003cstrong\u003ed\u003c/strong\u003e, Timeline for a single i.p. injection of ETP69 or vehicle in AD\u003csup\u003e+\u003c/sup\u003e and WT mice (10 DMSO-WT, 10 ETP69-WT, 7 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 8 ETP69-AD\u003csup\u003e+\u003c/sup\u003e). \u003cstrong\u003ee\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003ePercentage of alternations in the colour X-maze. \u003cstrong\u003ef\u003c/strong\u003e, Freezing time in the contextual fear conditioning. \u003cstrong\u003eg\u003c/strong\u003e, Immunofluorescence images and quantifications of H3K9me3 (red, μm\u003csup\u003e2\u003c/sup\u003e), 6E10-Aβ plaques (white, μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e), and GFAP-astrocytes (green, μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e) IR in the cortex (7 DMSO-WT, 7 ETP69-WT, 5 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 6 ETP69-AD\u003csup\u003e+\u003c/sup\u003e). Scale bar, 50 μm. \u003cstrong\u003eh–j\u003c/strong\u003e, Immunofluorescence images and quantification of H3K9me3 IR (red; % area) in different cell types (5 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 6 ETP69-AD\u003csup\u003e+\u003c/sup\u003e): NeuN-neurons (green) (\u003cstrong\u003eh\u003c/strong\u003e); GFAP-astrocytes (green) (\u003cstrong\u003ei\u003c/strong\u003e);\u003cstrong\u003e \u003c/strong\u003eIBA1 (white)/CD45 (green)-myelomonocytic cells (\u003cstrong\u003ej\u003c/strong\u003e). Scale bar, 20 μm. \u003cstrong\u003ek\u003c/strong\u003e, Quantification of activated microglia-associated IBA1 IR (μm\u003csup\u003e2\u003c/sup\u003e) in the cortex of AD\u003csup\u003e+\u003c/sup\u003e mice. \u003cstrong\u003el\u003c/strong\u003e, Pearson’s correlation between H3K9me3 and IBA1 IR. \u003cstrong\u003em\u003c/strong\u003e, Immunofluorescence images of CD45\u003csup\u003ehi \u003c/sup\u003e(green), IBA1 (white) (green) and 6E10-Aβ plaques (pink) and ratio of CD45\u003csup\u003ehi\u003c/sup\u003e-infiltrating monocyte-derived macrophages to IBA1 IR in the vicinity of 6E10-Aβ plaques. \u003cstrong\u003en\u003c/strong\u003e, Pearson’s correlation between CD45\u003csup\u003ehi\u003c/sup\u003e/IBA1 ratio and 6E10-Aβ plaque burden. \u003cstrong\u003eo\u003c/strong\u003e, Blood monocyte counts in 18-month-old AD\u003csup\u003e+\u003c/sup\u003e mice after ETP69 or vehicle administration (3 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 5 ETP69-AD\u003csup\u003e+\u003c/sup\u003e). D = DMSO, E = ETP69. Individual data points are presented with group means ± SEMs. Filled and empty circles represent male and female, respectively. Median, lower and upper quartiles are indicated on violin plots. ## p \u0026lt; 0.01, ### p \u0026lt; 0.001, and #### p \u0026lt; 0.0001: DMSO-AD\u003csup\u003e+\u003c/sup\u003e versus DMSO-WT mice; * p \u0026lt; 0.05, ** p\u003cem\u003e \u003c/em\u003e\u0026lt; 0.01, *** p \u0026lt; 0.001, and **** p \u0026lt; 0.0001: ETP69- versus DMSO- mice; by one- or two-way ANOVA followed by\u0026nbsp;Fisher’s LSD \u003cem\u003epost hoc\u003c/em\u003e test, or two-tailed unpaired Student’s t-test.\u003c/p\u003e","description":"","filename":"Fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-8913130/v1/ebc59d9700962ba3d2fba536.png"},{"id":104781191,"identity":"18d3e97a-8787-44ba-9b45-329cf59b6178","added_by":"auto","created_at":"2026-03-17 07:55:05","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":995966,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProteome signatures of ETP69 treatment in the brain of WT and AD\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e+\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e mice.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e, Venn diagram showing the overlap of all significant differentially expressed proteins (DEPs) in 14-month-old AD\u003csup\u003e+\u003c/sup\u003e versus WT mice (8 DMSO-WT, 7 ETP69-WT, 6 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 6 ETP69-AD\u003csup\u003e+\u003c/sup\u003e). \u003cstrong\u003eb\u003c/strong\u003e, Significant upregulated and downregulated DEPs. Blue and red numbers represent DEPs with |FC| above 1.2. \u003cstrong\u003ec\u003c/strong\u003e, Heatmaps displaying the hierarchical clustering of significant downregulated (blue) and upregulated (red) DEPs. \u003cstrong\u003ed\u003c/strong\u003e, Principal component analysis of protein expression profiles. DEPs significant in at least one of three comparisons (AD\u003csup\u003e+\u003c/sup\u003e ETP69/DMSO, WT ETP69/DMSO or AD\u003csup\u003e+\u003c/sup\u003e/WT) are included (total: 1118 proteins). \u003cstrong\u003ee\u003c/strong\u003e, Inverse relationship of DEP expression between AD\u003csup\u003e+\u003c/sup\u003e ETP69/DMSO and AD\u003csup\u003e+\u003c/sup\u003e/WT \u003cstrong\u003ef\u003c/strong\u003e, Volcano plot of all DEPs in ETP69-AD\u003csup\u003e+\u003c/sup\u003e mice (versus control AD\u003csup\u003e+\u003c/sup\u003e mice). « Stars indicate proteins that were further analysed. \u003cstrong\u003eg\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eAPP activation (z score = 2.39) in AD\u003csup\u003e+\u003c/sup\u003e (versus WT, green bars) mice using Ingenuity Pathway Analysis (IPA). All changes in protein expression were significant (green dots). DEPs in ETP69-treated AD\u003csup\u003e+\u003c/sup\u003e mice (versus control AD\u003csup\u003e+\u003c/sup\u003e mice) are shown (blue line) with significant DEPs indicated by blue dots. \u0026nbsp;\u003cstrong\u003eh\u003c/strong\u003e, Quantifications of MT1 and MT2 protein expression in 14-month-old mice by MS. \u003cstrong\u003ei\u003c/strong\u003e, Immunofluorescence images and quantification of MT2 (yellow, μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e) in the hippocampus of 18-month-old AD\u003csup\u003e+\u003c/sup\u003e mice (6 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 6 ETP69-AD\u003csup\u003e+\u003c/sup\u003e). Scale bar, 100 μm. DAPI nuclei (blue). \u003cstrong\u003ej,k\u003c/strong\u003e, Quantifications of cerebral (\u003cstrong\u003ej\u003c/strong\u003e) and plasma (\u003cstrong\u003ek\u003c/strong\u003e) HP and HPX in 14-month-old mice by MS and ELISA. \u003cstrong\u003el\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eProtein association network showing select protein roles in Aβ clearance, phagocytosis, and chemotaxis using the STRING database (v12.0). Nodes (proteins) are colour-coded according to function. The thickness of the edges indicates the confidence of the interaction from medium (0.4) to highest confidence (0.9). \u003cstrong\u003em\u003c/strong\u003e, Quantifications of\u003cstrong\u003e \u003c/strong\u003eICAM1, FCGR2B, complement components CFH and C3, CYP51, and LRPAP1 protein expression by MS. Individual data points are presented with group means ± SEMs. # p \u0026lt; 0.05, ## p \u0026lt; 0.01, and ### p \u0026lt; 0.001: DMSO-AD\u003csup\u003e+\u003c/sup\u003e versus DMSO-WT mice; * p \u0026lt; 0.05, ** p\u003cem\u003e \u003c/em\u003e\u0026lt; 0.01, *** p \u0026lt; 0.001, and **** p \u0026lt; 0.0001: ETP69 versus DMSO mice; by one-way ANOVA followed by\u0026nbsp;Fisher’s LSD \u003cem\u003epost hoc\u003c/em\u003e test or two-tailed unpaired Student’s t-test.\u003c/p\u003e","description":"","filename":"Fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-8913130/v1/bb0a89fa191aabefac6b2692.png"},{"id":104538706,"identity":"a32d8644-af5f-43af-8581-e61aa1c469b7","added_by":"auto","created_at":"2026-03-13 04:58:06","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1412675,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEnriched molecular pathways related to neuronal survival, synaptic plasticity, and cognition due to ETP69.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea–c\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eHeatmaps comparing activation z-scores for biological processes associated with neuroplasticity (\u003cstrong\u003ea\u003c/strong\u003e) and behaviour (\u003cstrong\u003eb\u003c/strong\u003e), and for upstream regulators (\u003cstrong\u003ec\u003c/strong\u003e) in 14-month-old AD\u003csup\u003e+\u003c/sup\u003e (versus WT) mice and in ETP69 (versus DMSO) AD\u003csup\u003e+\u003c/sup\u003e mice according to IPA. \u003cstrong\u003ed\u003c/strong\u003e, IPA diagram of proteins related to the upstream regulator BDNF (z-score = 2.45). The observed FC of each target in ETP69 (versus DMSO) AD\u003csup\u003e+\u003c/sup\u003e mice is indicated. \u003cstrong\u003ee\u003c/strong\u003e, Immunofluorescence images and quantifications of PLTP (red, μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e) and ANXA2 (green, μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e) IR in the hippocampus of 18-month-old AD\u003csup\u003e+\u003c/sup\u003e mice (4 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 6 ETP69-AD\u003csup\u003e+\u003c/sup\u003e). Scale bar, 100 μm.\u003cstrong\u003e \u003c/strong\u003eDAPI nuclei (blue). \u003cstrong\u003ef\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eQuantifications of NTRK2 and PCSK1 protein expression by MS. \u003cstrong\u003eg\u003c/strong\u003e, Members of the extended granin family differentially expressed in ETP69-AD\u003csup\u003e+\u003c/sup\u003e versus DMSO-AD\u003csup\u003e+\u003c/sup\u003e mice (green line) and in AD\u003csup\u003e+\u003c/sup\u003e versus WT mice (blue line). \u003cstrong\u003eh,i,\u003c/strong\u003e Western blot images (\u003cstrong\u003eh\u003c/strong\u003e) and normalised quantifications (\u003cstrong\u003ei\u003c/strong\u003e) of H3K9me3/total H3, VGF/β-actin, BDNF/GAPDH, and 12F4-Aβ expression in the hippocampus of 18-month-old mice (n = 3 per group). \u003cstrong\u003ej\u003c/strong\u003e, Immunofluorescence images and quantifications of BDNF (green, μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e),\u003cstrong\u003e \u003c/strong\u003eVGF (red, μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e) and ThioS-Aβ (white, μm\u003csup\u003e2\u003c/sup\u003e×10\u003csup\u003e3\u003c/sup\u003e) in the hippocampus and cortex of 18-month-old WT and AD\u003csup\u003e+\u003c/sup\u003e mice (5 DMSO-WT, 4 ETP69-WT, 5 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 6 ETP69-AD\u003csup\u003e+\u003c/sup\u003e). Scale bar, 100 μm. \u003cstrong\u003ek\u003c/strong\u003e, Pearson’s correlations between\u003cstrong\u003e \u003c/strong\u003ehippocampal and cortical H3K9me3 and VGF IR, between hippocampal H3K9me3/total H3 and VGF/β-actin, and between cortical VGF IR and memory retention performance in the Barnes maze. Individual data points are presented with group means ± SEMs. ### p \u0026lt; 0.001 and #### p \u0026lt; 0.0001: DMSO-AD\u003csup\u003e+\u003c/sup\u003e versus DMSO-WT mice; * p \u0026lt; 0.05, ** p\u003cem\u003e \u003c/em\u003e\u0026lt; 0.01, *** p \u0026lt; 0.001, and **** p \u0026lt; 0.0001: ETP69 versus DMSO mice; by one-way ANOVA followed by\u0026nbsp;Fisher’s LSD \u003cem\u003epost hoc\u003c/em\u003e test or two-tailed unpaired Student’s t-test.\u003c/p\u003e","description":"","filename":"Fig6.png","url":"https://assets-eu.researchsquare.com/files/rs-8913130/v1/23a72f2f0dab6178fde4bd3d.png"},{"id":104538707,"identity":"6eab3734-ba5c-43d9-bbf6-640c6d65302f","added_by":"auto","created_at":"2026-03-13 04:58:06","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":695845,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProteome signatures of ETP69 treatment in the retina of WT and AD\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e+\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e mice. a\u003c/strong\u003e, Venn diagram showing the overlap of significant DEPs in 14-month-old AD\u003csup\u003e+\u003c/sup\u003e mice versus WT mice (7 DMSO-WT, 7 ETP69-WT, 5 DMSO-AD\u003csup\u003e+\u003c/sup\u003e, 6 ETP69-AD\u003csup\u003e+\u003c/sup\u003e). \u003cstrong\u003eb\u003c/strong\u003e, Heatmap displaying the hierarchical clustering of significant downregulated (blue) and upregulated (red) DEPs between ETP69-AD\u003csup\u003e+\u003c/sup\u003e and DMSO-AD\u003csup\u003e+\u003c/sup\u003e mice. \u003cstrong\u003ec\u003c/strong\u003e, Inverse relationship of the DEP fold-change (FC) for ETP69- versus DMSO-AD\u003csup\u003e+ \u003c/sup\u003emice and for AD\u003csup\u003e+\u003c/sup\u003e versus WT mice. \u003cstrong\u003ed\u003c/strong\u003e, Volcano plot of all DEPs in the retina of ETP69-AD\u003csup\u003e+\u003c/sup\u003e mice (versus control AD\u003csup\u003e+\u003c/sup\u003e mice). \u003cstrong\u003ee\u003c/strong\u003e, GO enrichment analysis of all DEPs in at least 1 out of AD\u003csup\u003e+\u003c/sup\u003e/WT or AD\u003csup\u003e+\u003c/sup\u003e ETP69/DMSO comparisons (330 proteins). \u003cstrong\u003ef, \u003c/strong\u003eHeatmaps comparing IPA activation z scores for upstream regulators in the retina of AD\u003csup\u003e+\u003c/sup\u003e (versus WT) mice and ETP69- (versus DMSO) AD\u003csup\u003e+\u003c/sup\u003e mice. \u003cstrong\u003eg\u003c/strong\u003e, Expression profiles of select first and second neighbours upstream and downstream of SUV39H1 predicting its inhibition in the retina of ETP69-AD\u003csup\u003e+\u003c/sup\u003e mice, according to IPA. \u003cstrong\u003eh\u003c/strong\u003e, Members of the extended granin family differentially expressed in ETP69-AD\u003csup\u003e+\u003c/sup\u003e versus DMSO-AD\u003csup\u003e+\u003c/sup\u003e mice (green line) and in AD\u003csup\u003e+\u003c/sup\u003e versus WT mice (blue line). \u003cstrong\u003ei\u003c/strong\u003e, Analysis of successful alternations in the colour X-maze, according to arm sequence and colour of the starting arm in 14-month-old mice. Group means + SEMs are shown. # p \u0026lt; 0.05 and ## p \u0026lt; 0.01: DMSO-AD\u003csup\u003e+\u003c/sup\u003e mice versus DMSO-WT mice; * p \u0026lt; 0.05, *** p \u0026lt; 0.001: ETP69- versus DMSO- mice; by one-way ANOVA followed by\u0026nbsp;Fisher’s LSD \u003cem\u003epost hoc\u003c/em\u003e test.\u003c/p\u003e","description":"","filename":"Fig7.png","url":"https://assets-eu.researchsquare.com/files/rs-8913130/v1/52872dcb79d1f77b26b9f35a.png"},{"id":104784496,"identity":"3129d751-4ff6-483c-89a6-46f7bb6bfca8","added_by":"auto","created_at":"2026-03-17 08:08:00","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":9849372,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8913130/v1/22a22dc8-ea65-4322-b9ab-d29a3187e043.pdf"},{"id":104538710,"identity":"03d8e02c-3507-43dc-8e15-11184a17f64f","added_by":"auto","created_at":"2026-03-13 04:58:06","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":10541503,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExtended Data Figures and Tables\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eExtended Data Figures 1 to 14\u003c/p\u003e\n\u003cp\u003eExtended Data Tables 1 to 14\u003c/p\u003e","description":"","filename":"FuchsVitRentsendorjetalExtendedDataFile3.11.2026F.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8913130/v1/e3646c71f2fd642c02372786.pdf"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nKLB is the Co-Chairman and shareholder, and MKH is a scientific advisor, of Fortem Neurosciences, Inc. Unrelated to this study: YK, KLB and MKH are co-founders of NeuroVision Imaging, Inc. The other authors have no conflicts to disclose.","formattedTitle":"H3K9me3 inhibition reverses Alzheimer′s progression by restoring synaptic and immune proteostasis across the brain–retina axis","fulltext":[{"header":"Main","content":"\u003cp\u003eEpigenetic processes are increasingly recognized as key contributors to ageing and neurodegenerative disorders, including Alzheimer\u0026rsquo;s disease (AD)\u003csup\u003e1-8\u003c/sup\u003e, the most common cause of senile dementia\u003csup\u003e9\u003c/sup\u003e. Canonical neuropathological hallmarks of AD, amyloid \u0026beta;-protein (A\u0026beta;) plaques and neurofibrillary tangles composed of hyperphosphorylated tau, together with neuroinflammation, vascular injury, and progressive synaptic and neuronal loss\u003csup\u003e10-12\u003c/sup\u003e, extend beyond the brain to the neurosensory retina\u003csup\u003e13-27\u003c/sup\u003e. Although\u0026nbsp;retinal epigenome undergoes dynamic remodelling with age and disease\u003csup\u003e28-31\u003c/sup\u003e, epigenetic mechanisms underlying AD-associated retinal dysfunction remain largely uncharacterized. Because the retina is an anatomically accessible extension of the central nervous system (CNS), defining epigenetic signatures that track AD progression at the brain\u0026ndash;retina interface could enable noninvasive biomarker development for longitudinal monitoring and inform precision epigenetic interventions.\u003c/p\u003e\n\u003cp\u003eHistone methylation and acetylation are central regulators of gene expression across the lifespan\u003csup\u003e32,33\u003c/sup\u003e and essential for synaptic plasticity as well as learning and memory formation\u003csup\u003e34-37\u003c/sup\u003e. Among these modifications, repressive chromatin marks\u0026mdash;di- and tri-methylation of histone H3 lysine 9 (H3K9me2/3)\u0026mdash;are increased in AD patient brains and in AD mouse models\u003csup\u003e38-40\u003c/sup\u003e. Notably, elevated H3K9me3, catalysed by the histone methyltransferase SUV39H1, is linked to transcriptional silencing of genes critical for synaptic function\u003csup\u003e39\u003c/sup\u003e. Given the reversibility of epigenetic modifications, histone-modifying enzymes represent promising therapeutic targets for preserving neuronal function during ageing and neurodegeneration\u003csup\u003e41,42\u003c/sup\u003e.\u0026nbsp;However, whether targeting H3K9me3 can mitigate AD-related dysfunction in the brain and retina, and thereby preserve cognitive and visual function, remains unknown.\u003c/p\u003e\n\u003cp\u003eIn this study, we profiled H3K9me3 and related proteomic pathways in postmortem brains and corresponding retinas of cognitively normal (CN) individuals and patients with mild cognitive impairment (MCI due to AD) or AD dementia. We found a strong association between retinal and brain H3K9me3 levels, AD pathology, and cognitive status\u0026mdash; starting to elevate at the earliest clinical stage (MCI). To define the functional contribution of this repressive mark, we pharmacologically inhibited the H3K9 methyltransferase SUV39H1, using the synthetic epidithiodiketopiperazine compound ETP69 (NT1721)\u003csup\u003e43\u003c/sup\u003e, in two transgenic mouse models of AD (APPswe/PS1dE9 and APPswe/tauP301L/PS1\u003cem\u003e\u003csup\u003etm1Mpm\u003c/sup\u003e\u003c/em\u003e). By integrating brain\u0026ndash;retina biochemical and proteomic profiling, quantitative histology, blood-based immunophenotyping, and behavioural analyses, pharmacological inhibition of SUV39H1 reduced H3K9me3 burden, attenuated AD-like neuropathology, restored synaptic integrity, and rescued both cognitive performance and visual function. Mechanistically, ETP69 conferred neuroprotection by re-establishing retina and brain proteostasis, accompanied by coordinated remodelling of pathways governing innate immune activation and trophic support, including BDNF\u0026ndash;VGF\u0026ndash;granin signalling.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eElevated cerebral and retinal H3K9me3 associates with neuropathology and cognitive impairment in MCI/AD\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eH3K9me3 levels in the brain and retina were measured in patients with a clinical and post-mortem neuropathological diagnosis of AD dementia (n=13, mean age 85.5 ± 2.7 years, 7 females/6 males) or MCI (due to AD; n=7, mean age 91.1 ± 1.9 years, 4 females/3 males), and compared to age- and sex-matched CN individuals (n=6, mean age 91.3 ± 2.8 years, 4 females/2 males; human donor information is summarized in Table 1 and Extended Data Table 1). In the dorsolateral prefrontal cortex (Brodmann area 9; A9), a region critical for executive function and affected in AD, immunohistochemical (IHC) quantification of H3K9me3 immunoreactive (IR) area normalised to DAPI nuclear density revealed robust increases in both disease groups, with significant 3.3-fold and 2.2-fold elevations in AD and MCI, respectively, relative to CN controls (AD, p \u0026lt; 0.0001; MCI, p = 0.0207; Fig. 1a,b; H3K9me3 raw area analysis in Extended Data Fig. 1a). No difference between male and female patients was detected (Extended Data Fig. 1b). Higher cortical H3K9me3 strongly correlated with cognitive deficits, with higher CDR) and lower MMSE scores (CDR: \u003cem\u003er\u003c/em\u003e = 0.68, p = 0.0004; MMSE: \u003cem\u003er\u003c/em\u003e = –0.82, p \u0026lt; 0.0001; Fig. 1c,d). In the cortical A9 region, H3K9me3 levels more strongly correlated with NFT severity (\u003cem\u003er\u003c/em\u003e = 0.75, p = 0.0001; Fig. 1e) than Aβ-plaque burden (\u003cem\u003er\u003c/em\u003e = 0.57, p = 0.0070; Fig. 1f). Cortical H3K9me3 further associated with AD neuropathological indices, disease stage, and cognitive performance across analyses stratified as CN+MCI, MCI+AD, or all groups combined (heatmap summary in Extended Data Fig. 1c).\u003c/p\u003e\n\u003cp\u003eIn the retina from the same cohort, quantitative histology of H3K9me3 per nuclei showed significant 2.2- and 1.6-fold increases in AD and MCI patients, respectively, compared to CN individuals (p \u0026lt; 0.0001 and p = 0.0087; Fig. 1g,h; H3K9me3 raw area analysis in Extended Data Fig. 1d), with no difference between male and female AD patients (Extended Data Fig. 1e). H3K9me3 immunoreactivity was detected across all retinal nuclear layers, with particularly strong signal in the retinal ganglion cell layer and inner nuclear layer, and enrichment in cells proximal to Aβ deposits, most evident in MCI and AD retinas (Fig. 1g). Elevated retinal H3K9me3 closely tracked retinal 4G8\u003csup\u003e+\u003c/sup\u003e Aβ deposition (r = 0.85, p \u0026lt; 0.0001) and, albeit to a lesser extent, remained strongly associated with retinal T22\u003csup\u003e+\u003c/sup\u003e tau oligomers (r = 0.65, p = 0.0051; Fig. 1i). Retinal H3K9me3 also correlated with retinal ionized calcium-binding adaptor molecule 1 (IBA1)\u003csup\u003e+\u003c/sup\u003e microgliosis and glial fibrillary acidic protein (GFAP)\u003csup\u003e+\u003c/sup\u003e macrogliosis (\u003cem\u003er\u003c/em\u003e = 0.64, p = 0.0046 and \u003cem\u003er\u003c/em\u003e = 0.62, p = 0.0102, respectively; Extended Data Fig. 1f). Notably, retinal H3K9me3 tightly correlated with cortical H3K9me3 (\u003cem\u003er\u003c/em\u003e = 0.70, p = 0.0004; Fig.1j), and associated with cognitive status (CDR: \u003cem\u003er\u003c/em\u003e = 0.65, p = 0.0008 and MMSE: \u003cem\u003er\u003c/em\u003e = –0.62, p = 0.0042; Fig. 1k). These data suggest a coordinated brain–retina increase in the repressive chromatin mark H3K9me3 that tracks AD neuropathology, disease stage, and cognitive impairment.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eProteomics identifies SUV39H1–H3K9 hypertrimethylation in AD brain and retina\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo extend our histological findings and define perturbations in epigenetic pathways across the AD brain and retina, with a focus on H3K9 regulation, we performed a targeted reanalysis of mass spectrometry–based proteomic datasets from independent human cohorts \u003csup\u003e16\u003c/sup\u003e (temporal cortex: n=10 AD, n=8 CN; temporal hemiretina: n=6 AD, n=6 CN; Fig. 1l–o). Metascape Gene Ontology analysis of differentially expressed proteins (DEPs) in AD brains, revealed significant enrichment of pathways governing epigenetic regulation, chromatin organization, nucleosome assembly, and histone modification (Fig. 1l). Importantly, ingenuity pathway analysis (IPA) predicted SUV39H1 activation, by expression profiles of directly and indirectly connected DEP network in AD cortices (Fig. 1m), consistent with elevated cortical H3K9me3 observed in MCI and AD brains. Histone variants were broadly increased, including H1, which stabilizes linker DNA and promotes higher-order chromatin compaction, H3-3, which increases with ageing\u003csup\u003e44-46\u003c/sup\u003e, and H2AZ1, a transcriptional repressor of activity-dependent plasticity genes\u003csup\u003e47\u003c/sup\u003e (Extended Data Fig. 2a,b). Enzymes regulating histone arginine methylation were also altered: PRMT1, which methylates H4R3 and supports H3K9 acetylation, was reduced, whereas PRMT5, which methylates H3R8 and antagonizes H3K9 acetylation, was increased, collectively shifting chromatin toward reduced H3K9ac and enhanced H3K9 methylation (Extended Data Fig. 2a,b).\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Proteomic profiling of AD retinas similarly revealed dysregulation of pathways linked to epigenetic control, chromatin organization, heterochromatin formation, and histone H3 methyltransferase activity (Fig. 1n and Extended Data Fig. 2c). The expression pattern of histone modifying enzymes in AD retina predicted enhanced H3K9 hypermethylation (Fig. 1o). Lysine acetyltransferase 2B (KAT2B) and\u0026nbsp;TATA-box binding protein associated factor 1 (TAF1), which acetylate H3K9, and the H3K9me3-specific lysine demethylases 4B and 4C (KDM4B/C), were downregulated, collectively promoting H3K9 trimethylation. Lysine acetyltransferase 7 (KAT7), which restricts SUV39H1-mediated heterochromatin spreading\u003csup\u003e48\u003c/sup\u003e, was also reduced, whereas the arginine demethylase and lysine hydroxylase Jumonji domain containing 6 (JMJD6), which demethylates H4R3 \u003csup\u003e49\u003c/sup\u003e and can bias chromatin toward H3K9 methylation, was increased. Together with the histological findings, these proteomic signatures converge on elevated H3K9me3 in both AD brain and retina, supporting H3K9me3 as a cross-compartment epigenetic correlate of disease progression and providing a mechanistic rationale for targeting SUV39H1-heterochromatinization as a therapeutic strategy in AD.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSUV39H1 inhibition rescues cognitive and visual function in aged AD models\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe next tested whether ETP69-mediated SUV39H1 inhibition improves cognitive and visual performance in 18-month-old APPswe/PS1dE9 (AD⁺) mice and age- and sex-matched WT littermates (Fig. 2a–h and Extended Data Figs. 3,4). Mice received intraperitoneal ETP69 (10 mg kg⁻¹) under distinct dosing regimens (n = 118), whereas controls received vehicle (DMSO), and all animals completed a behavioural battery spanning locomotion, hippocampus-dependent spatial learning and memory, and visual stimulus discrimination to capture both rapid and sustained effects from 1 to 14 days post-treatment (Fig. 2a,b and Extended Data Figs. 3a,4a). Short-term testing (days 1–3; open field, Y-maze, and colour- and contrast-mode visual X-maze) showed that ETP69 did not alter locomotor activity in either genotype (distance travelled, resting time, average speed, rearing, or arm entries; Fig. 2c and Extended Data Fig. 3b,c). Notably, ETP69 reversed working-memory deficits in aged AD⁺ mice, increasing spontaneous alternations across assays (Y-maze: p = 0.0093; colour X-maze: p \u0026lt; 0.0001; contrast X-maze: p = 0.0020; Fig. 2d; repeated-dose data in Extended Data Fig. 3d). A single dose also increased alternations in aged WT mice in the colour-mode X-maze (p = 0.0027; Fig. 2d). Repeated ETP69 dosing rescued colour-mode X-maze visual performance in AD⁺ mice, restoring blue-arm entries and blue↔white bidirectional transitions to WT-like levels (baseline deficits: p = 0.0059 and p = 0.0009; rescue: p = 0.0383 and p = 0.0035; Fig. 2e and Extended Data Fig. 3e,f).\u003c/p\u003e\n\u003cp\u003eWe further examined whether ETP69 elicited sustained improvements in hippocampus-dependent learning and memory from days 4–14 using the Barnes maze and contextual fear conditioning (combined data in Fig. 2f–h; regimen-specific data in Extended Data Fig. 4). In the Barnes maze, vehicle-treated AD⁺ mice made more errors than WT across all phases (acquisition: repeated measures (RM) ANOVA, F(1,30) = 27.40, p \u0026lt; 0.0001; retention: p \u0026lt; 0.0001; reversal: F(1,23) = 34.07, p \u0026lt; 0.0001), whereas ETP69 reduced errors during acquisition (F(1,27) = 10.23, p = 0.0035), retention (p \u0026lt; 0.0001), and reversal (F(1,20) = 16.19, p = 0.0007), restoring performance to near WT levels (Fig. 2f and Extended Data Fig. 4b). Movement-transition chord diagrams further supported this rescue, showing focused search near the escape box in ETP69-treated AD⁺ mice during retention (red ribbons) and reversal (blue ribbons), in contrast to the non-goal-oriented exploration of vehicle-treated AD⁺ mice (Fig. 2g).\u003c/p\u003e\n\u003cp\u003eAssociative learning and memory assessed by contextual fear conditioning showed increased freezing 24 h after conditioning in both genotypes following ETP69 treatment (WT: p = 0.0119 in the first minute; AD⁺: RM ANOVA, F(1,26) = 11.71, p = 0.0021; Fig. 2h and Extended Data Fig. 4c,d). Overall, ETP69 produced rapid and sustained improvements in spatial learning, long-term memory retention, reversal learning, colour vision, and associative memory in aged AD⁺ mice. Performance did not differ by sex across Y-maze, colour- and contrast-mode X-maze, and Barnes maze (Extended Data Fig. 4e); accordingly, males and females were pooled for subsequent molecular analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCerebral H3K9me3 reduction curbs Aβ pathology and astrogliosis while preserving synaptic integrity\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGiven the behavioural benefits of SUV39H1 inhibition, we next examined its impact on AD-associated neuropathology. Immunohistochemistry and immunoblotting of 18-month-old AD\u003csup\u003e+\u003c/sup\u003e mouse brains showed elevated H3K9me3 versus WT littermates, with a 1.3–1.5-fold increase in cortex (p \u0026lt; 0.0001; Fig. 2i), including layers II/III and VI (Extended Data Fig. 5a,b), and higher levels across the hippocampal formation (p = 0.0782 to p \u0026lt; 0.0001; Fig. 2j,k), spanning CA1–CA3 and the dentate gyrus (DG) (Extended Data Fig. 5c). A single ETP69 injection markedly lowered H3K9me3 in AD\u003csup\u003e+\u003c/sup\u003e mice as early as 4 days post-injection, reducing levels by 40% in cortex (p \u0026lt; 0.0001; Fig. 2i) and by 32% in hippocampus (p = 0.0454; Fig. 2j) to below WT control levels. These effects persisted 15 days post-injection, with 40% reduction in cortex (p= 0.0069) and 47% in hippocampus (CA: 54% and DG 42%, p\u0026lt; 0.0001; Fig. 2i,k and Extended Data Fig. 5c). In aged WT mice, ETP69 also significantly lowered H3K9me3 by 27% in cortex (p = 0.0004; Fig. 2i) and 36% in hippocampus at day 4 (p = 0.0286; Fig. 2j), and 36% in the CA region at day 15 (p= 0.0185; Extended Data Fig. 5c).\u003c/p\u003e\n\u003cp\u003eEvaluation of core AD pathology and associated gliosis indicated attenuation of Aβ plaque burden and GFAP reactive astrogliosis following ETP69 treatment. In cortex, ETP69 caused marked reductions in GFAP\u003csup\u003e+\u003c/sup\u003e and 6E10-Aβ plaques at both 4 days (GFAP: 58%, p \u0026lt; 0.0001; 6E10: 42%, p = 0.0004) and 15 days post-injection (GFAP: 44%, p = 0.0133; 6E10: 66%, p = 0.0004; Fig. 2l). In hippocampus, while vehicle-treated AD\u003csup\u003e+\u003c/sup\u003e mice exhibited pronounced astrogliosis compared to WT mice (3.3-fold, p \u0026lt; 0.0001), ETP69 significantly attenuated GFAP levels 4 days after (29%, p = 0.0223; Fig. 2m) and remained reduced 15 days post-treatment (27%, p = 0.0318), accompanied by a 33% reduction in 6E10-Aβ plaque (p = 0.0066; Fig. 2n).\u003c/p\u003e\n\u003cp\u003eConsistent with our findings in AD patients, elevated cerebral H3K9me3 levels in AD-model mice strongly associated with cognitive deficits and neuropathology (Fig 2o and Extended Data Fig. 5d). On day 4, higher cortical H3K9me3 levels correlated with poorer performance in the colour- and contrast-mode X-maze (\u003cem\u003er\u0026nbsp;\u003c/em\u003e= −0.77, p\u0026lt; 0.0001 and \u003cem\u003er\u0026nbsp;\u003c/em\u003e= −0.48, p= 0.0416, respectively). On day 15, elevated cortical and hippocampal H3K9me3 levels correlated with increased errors in the Barnes maze (CC: retention,\u003cem\u003e\u0026nbsp;r\u0026nbsp;\u003c/em\u003e= 0.77, p= 0.0053; Hipp: reversal, \u003cem\u003er\u0026nbsp;\u003c/em\u003e= 0.57, p= 0.0113 and retention,\u003cem\u003e\u0026nbsp;r\u0026nbsp;\u003c/em\u003e= 0.49, p= 0.0299) and reduced freezing in the fear conditioning (Hipp: \u003cem\u003er\u0026nbsp;\u003c/em\u003e= −0.63, p= 0.0063). Notably, H3K9me3 levels strongly associated with Aβ plaque burden in both cortex (day 4: \u003cem\u003er\u0026nbsp;\u003c/em\u003e= 0.79, p= 0.0020; day 15: \u003cem\u003er\u0026nbsp;\u003c/em\u003e= 0.84, p= 0.0011) and hippocampus (day 15: \u003cem\u003er\u0026nbsp;\u003c/em\u003e= 0.82, p= 0.0033), and with astrogliosis (day 4: \u003cem\u003er\u0026nbsp;\u003c/em\u003e= 0.72, p= 0.0002) in the cortex. Collectively, these findings show that a single dose of ETP69 drives a sustained reduction of cerebral H3K9me3, astrogliosis, and Aβ pathology in aged AD-model mice.\u003c/p\u003e\n\u003cp\u003eSince dendritic spine loss is a strong predictor of cognitive decline \u003csup\u003e50-52\u003c/sup\u003e, we asked whether cerebral H3K9me3 reduction preserves synaptic architecture in the context of AD. We therefore quantified dendritic spine structure and integrity by Golgi–Cox staining in brains from aged ETP69-treated AD\u003csup\u003e+\u003c/sup\u003e and WT mice (regimen B; Fig. 3a,b). Analysis of spine subtypes, showed reduced thin spine density (thin spines/μm of dendrite), reflecting newly formed synapses, in control AD\u003csup\u003e+\u003c/sup\u003e mice, with 28% reduction in the cortex (p = 0.0004) and 14% in the hippocampus (p = 0.0327; Fig. 3e). ETP69 substantially increased their density in both regions and genotypes (WT: CC, 1.2-fold, p = 0.0006; Hipp, 1.3-fold, p = 0.0013; AD\u003csup\u003e+\u003c/sup\u003e:CC, 1.6-fold, p \u0026lt; 0.0001; Hipp, 1.5-fold, p = 0.0003; Fig. 3e). In hippocampus, ETP69 also increased stubby spine density (2-fold, p = 0.0461; Fig. 3e) and reduced mushroom spines (54%, p = 0.0085; Extended Data Fig. 6c), with no significant change in the cortex. The overall spine number increased with ETP69 in both brain regions and genotypes (Fig. 3f).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eImportantly, increased thin spine density correlated with improved performance in the Barnes maze and contextual fear conditioning (\u003cem\u003er\u0026nbsp;\u003c/em\u003e= −0.54, p = 0.0301 and \u003cem\u003er\u0026nbsp;\u003c/em\u003e= 0.67, p = 0.0063 respectively; Fig. 3g). By quantifying spine subtype composition as a proportion of total spines, we found that thin spines were the predominant subtype in aged WT mice (60–70%) and remained dominant but were reduced in AD\u003csup\u003e+\u003c/sup\u003e mice (~50%; Fig. 3h and Extended Data Fig. 6d). ETP69 increased the thin-to-total spine ratio while decreasing the mushroom-to-total spine ratio (Extended Data Fig. 6e). Notably, Barnes maze performance inversely correlated with the thin-to-total spine ratio (r = −0.56, p = 0.0232; Fig. 3i). Finally, hippocampal synaptic proteins were increased in ETP69-treated AD⁺ mice, with higher postsynaptic density protein PSD95 (1.3-fold, p = 0.0137; Fig. 3j) and presynaptic synaptophysin SYP (1.9-fold, p = 0.0140; Fig. 3k). Together, these data suggest that lowering cerebral H3K9me3 restores synaptic integrity, underpinning epigenetic-based cognitive recovery in aged AD-model mice.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTargeting H3K9me3 attenuates tauopathy and amyloidosis, reshaping microglia and monocyte responses in midlife AD models\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the next set of experiments, we examined the effects of H3K9me3 reduction on tau and amyloid pathology, as well as cell-type–specific responses, in 14-month-old AD transgenic mice harbouring established amyloidosis and/or tauopathy, before the onset of ageing processes. In 3xTg-AD mice (APPSwe/tauP301L/PS1\u003csup\u003etm1Mpm\u003c/sup\u003e), which develop hyperphosphorylated tau and NFTs by the age of 12 months \u003csup\u003e53\u003c/sup\u003e, a single intraperitoneal injection of ETP69 (10 mgkg\u003csup\u003e-1\u003c/sup\u003e) significantly reduced cortical levels of H3K9me3 and paired helical filaments (PHF)-tau IR (62%, p = 0.0064 and 39%, p = 0.0003, respectively; Fig. 5a,b). Furthermore, elevated cortical H3K9me3 strongly correlated with PHF-tau accumulation (\u003cem\u003er\u003c/em\u003e = 0.78, p = 0.0050; Fig. 5c).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn 14-month-old APPswe/PS1dE9 (AD⁺) mice, a single intraperitoneal injection of ETP69 (10 mgkg\u003csup\u003e-1\u003c/sup\u003e) restored cognitive functions as observed in the colour X-maze and fear conditioning (Fig. 4d–f and Extended Data Fig. 7a–d). While locomotor activity, measured by total arm entries, was unaffected by ETP69 (Extended Data Fig. 7a), the cognitive deficits observed in AD\u003csup\u003e+\u003c/sup\u003e versus WT mice (X-maze: p = 0.0009; Fear conditioning: p = 0.0021) was reversed by ETP69, as shown by increased percentage of X-maze alternations (p = 0.0089; Fig. 4e) and prolonged freezing in the fear conditioning (p = 0.0203; Fig. 4f). ETP69 also enhanced cognitive performance in WT mice (Extended Data Fig. 7b–c). In contrast, AD-associated visual impairment, manifested by reduced B↔W bidirectional transitions (p = 0.0278), was not significantly rescued by the single ETP69 dose (Extended Data Fig. 7d).\u003c/p\u003e\n\u003cp\u003eTo determine whether cognitive improvement at this earlier disease stage was accompanied by underlying tissue and cellular remodelling, we performed complementary neuropathological analyses. Indeed, AD\u003csup\u003e+\u003c/sup\u003e mice exhibited elevated cortical H3K9me3 relative to WT controls (1.6-fold; p \u0026lt; 0.0001), and ETP69 treatment normalized H3K9me3 in AD\u003csup\u003e+\u003c/sup\u003e mice while further reducing H3K9me3 in WT mice (40–42% decrease; p \u0026lt; 0.0001; Fig. 4g and Extended Data Fig. 7e). Consistent with our findings in aged mice, ETP69 mitigated neuropathology in these younger AD\u003csup\u003e+\u003c/sup\u003e animal models, significantly reducing 6E10-Aβ plaque burden (40%, p = 0.0019; Fig. 4g) and GFAP astrogliosis (48%, p \u0026lt; 0.0001; Fig. 4g and Extended Data Fig. 7e,f). Pearson’s \u003cem\u003er\u003c/em\u003e correlation analyses further showed that lower cortical H3K9me3 was associated with improved cognitive performance as detected in the fear conditioning (\u003cem\u003er\u003c/em\u003e = −0.59, p = 0.0037), and colour-mode X-maze (\u003cem\u003er\u003c/em\u003e = −0.59, p = 0.0061), as well as with reduced Aβ plaque load (\u003cem\u003er\u003c/em\u003e = 0.78, p = 0.0042; Extended Data Fig. 7g).\u003c/p\u003e\n\u003cp\u003eTo define the cell-type–specific effects of ETP69 on H3K9me3, we performed co-immunolabelling for H3K9me3 with neuronal (NeuN; Fig. 4h), astrocytic (GFAP; Fig. 4i), and myeloid (IBA1\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e microglia and monocyte/macrophages; Fig. 4j) markers in the brains of these AD\u003csup\u003e+\u003c/sup\u003e murine models. H3K9me3 was preferentially lowered in myelomonocytes (69%, p = 0.0002) and neurons (45%, p \u0026lt; 0.0001), whereas astrocytes showed only a non-significant trend (18%, p = 0.0695). Building on this myeloid cell selectivity, we next interrogated the inflammatory milieu surrounding Aβ plaques. ETP69 treatment markedly reduced cortical IBA1 immunoreactivity (40% reduction; p = 0.0021; Fig. 4k and Extended Data Fig. 7h), and the extent of cortical H3K9me3 lowering closely correlated with diminished microgliosis (r = 0.69; p = 0.0182; Fig. 4l), linking H3K9me3 derepression to inflammatory response. Notably, ETP69 enhanced infiltration of monocyte‑derived macrophages (IBA1⁺CD45\u003csup\u003ehigh\u003c/sup\u003e) surrounding 6E10-Aβ plaques by 1.8‑fold (p = 0.0006; Fig. 4m and Extended Data Fig. 7i). In parallel with reduced plaque burden and microgliosis, ETP69 robustly shifted the cortical myeloid compartment towards peripheral infiltration, increasing the proportion of IBA1\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003ehigh\u003c/sup\u003e peripherally derived macrophages within total IBA1\u003csup\u003e+\u003c/sup\u003e cells by 3.2-fold compared to control AD\u003csup\u003e+\u003c/sup\u003e mice (p = 0.0004; Fig. 4m). The expanded brain infiltrating monocyte/macrophage‑covered area inversely correlated with cortical Aβ load (r = −0.69, p = 0.0173; Fig. 4n), consistent with their reported phagocytic role in Aβ clearance \u003csup\u003e54-61\u003c/sup\u003e. Peripheral immune profiling in a subset of 18‑month‑old AD⁺ mice further supported this mechanism: ETP69 markedly increased circulating monocytes (3.8‑fold, p = 0.0039) and granulocytes (2.9‑fold, p = 0.0375), without altering lymphocyte or red blood cell counts (Fig. 4o and Extended Data Fig. 7j). Together, these findings indicate that ETP69 preferentially reduces cerebral H3K9me3 within microglia and peripherally derived mononuclear phagocytes, and promotes their accumulation at Aβ plaque sites, thereby enhancing Aβ clearance and limiting neuroinflammation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLimiting H3K9me3 re-establishes brain proteostasis and strengthens immune pathways supporting Aβ clearance\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo delineate the molecular basis of neuroprotection conferred by SUV39H1 inhibition–mediated H3K9me3 lowering, we performed global MS-based proteomic profiling of brains from the same mouse cohort. Among 5097 proteins identified across all samples (Fig. 5a and Extended Data Fig. 8a), 4358 were detected in at least three animals per group, and 1119 DEPs (p \u0026lt; 0.05) distributed between the following three comparisons: AD\u003csup\u003e+\u003c/sup\u003e versus WT (AD\u003csup\u003e+\u003c/sup\u003e/WT: 747 DEPs), ETP69- versus vehicle-treated WT (WT ETP69/DMSO: 318 DEPs), and ETP69- versus vehicle-treated AD\u003csup\u003e+\u003c/sup\u003e mice (AD\u003csup\u003e+\u003c/sup\u003e ETP69/DMSO: 371 DEPs). Only 40 DEPs overlapped between WT ETP69/DMSO and AD\u003csup\u003e+\u003c/sup\u003e ETP69/DMSO, indicating genotype-specific proteomic responses to treatment consistent with distinct H3K9me3 landscape in healthy versus AD\u003csup\u003e+\u003c/sup\u003e brains (Fig. 5a). Overall, ETP69 elicited a balanced distribution of down- and upregulated DEPs; however, applying a \u0026gt;1.2-fold-change threshold revealed a pronounced shift toward upregulation in both genotypes (Fig. 5b; DEPs are listed in Extended Data Tables 2–7). Heatmaps and principal component analysis (PCA) showed clear clustering by genotype and treatment group (Fig. 5c,d and Extended Data Fig. 8b,c). Notably, of the 162 DEPs shared between the AD\u003csup\u003e+\u003c/sup\u003e/WT and AD\u003csup\u003e+\u003c/sup\u003e ETP69/DMSO comparisons, ETP69 reversed 159 AD-associated changes (R\u003csup\u003e2\u003c/sup\u003e = 0.86, p \u0026lt; 0.0001, Fig. 5e). With a more inclusive threshold (|FC| \u0026gt; 1.04), 1683 of 2005 shared proteins (84%) were similarly redirected toward WT-like levels (Extended Data Fig. 9). Gene Ontology enrichment analysis of these treatment-driven reversals revealed coordinated upregulation of pathways involved in cytoskeleton organization, synaptic function, and dendritic spine maintenance, accompanied by downregulation of AD-linked metabolic and lipid oxidation. PANTHER classification showed that metabolite-converting/protein-modifying enzymes, nucleic acid metabolism and translation, and membrane trafficking/transport were the most represented functional categories across comparisons (Extended Data Fig. 10a). Volcano plots highlight the top 20 up- and top 20 downregulated DEPs in the treatment versus control AD\u003csup\u003e+\u003c/sup\u003e brains (Fig. 5f; AD\u003csup\u003e+\u003c/sup\u003e/WT and WT ETP69/DMSO comparisons in Extended Data Fig. 10b).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo define the molecular networks targeted by lowering cerebral H3K9me3 in AD models, we next performed IPA. As expected, IPA predicted significant activation of the amyloid precursor protein (APP) in AD\u003csup\u003e+\u003c/sup\u003e compared to WT mice (z-score = 2.39, adj. p \u0026lt; 0.0090, Fig. 5g), including canonical AD markers, such as clusterin (CLU), apolipoprotein E (APOE), midkine (MDK), modulates APP processing and Aβ generation and the microglial activation marker galectin-3 (LGALS3); yet their expression was unchanged by this epigenetic-targeting drug. Nevertheless, 19 additional APP-linked proteins, including metallothionein 1 (MT1), lysosome-associated membrane glycoprotein 1 (LAMP1), ketimine reductase μ-crystallin (CRYM), microtubule-associated protein 6 (MAP6), ATP-binding cassette sub-family G member 1 (ABCG1), prostaglandin-H2 D-isomerase (PTGDS) and γ-aminobutyric acid type B receptor subunit 2 (GABBR2), shifted toward WT-like abundance in AD\u003csup\u003e+\u003c/sup\u003e brains after treatment, consistent with selective normalization of AD-associated proteomic dysregulation. Interestingly, microtubule-associated protein tau (MAPT), a major component of the neuronal cytoskeleton, was also a top downregulated protein in AD\u003csup\u003e+\u003c/sup\u003e versus WT mice and strongly upregulated in response to ETP69 in AD\u003csup\u003e+\u003c/sup\u003e mice (2.4-fold, p = 0.0188; Fig. 5g).\u003c/p\u003e\n\u003cp\u003eNotably, the stress- and inflammation-responsive proteins MT1 and MT2 were among the most strongly downregulated DEPs in ETP69-treated AD\u003csup\u003e+\u003c/sup\u003e mice (Fig. 5f,h). Reduced MT2 abundance by ETP69 was independently validated by IHC in aged AD\u003csup\u003e+\u003c/sup\u003e mice (Fig. 5i). In addition, the neuroprotective haemoglobin-scavenging proteins haptoglobin (HP) and hemopexin (HPX) were among the most strongly upregulated proteins in ETP69-treated WT and AD\u003csup\u003e+\u003c/sup\u003e mice (Fig. 5j). HPX was reduced in AD\u003csup\u003e+\u003c/sup\u003e versus WT mice, revealing a disease-associated deficit that ETP69 restored. Consistently, circulating HPX and HP increased markedly after treatment and strongly correlated with their cerebral abundance (Fig. 5k and Extended Data Fig. 11a,b), indicating systemic engagement of an early innate immune response.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn parallel to our blood and brain findings, proteomics from the same cohort showed that ETP69 enhanced pathways involved in Aβ clearance and innate immune activation (Fig. 5l,m and Extended Data Fig. 11c–e). Intercellular adhesion molecule 1 (ICAM1) and guanine nucleotide exchange factor 3 (VAV3), which facilitate cerebral monocyte recruitment and transendothelial migration, were induced in ETP69-treated brains. Additionally, the low affinity immunoglobulin gamma Fc region receptor II (FCGR2B), Rho GTPase-activating protein 25 (AHRGAP25), and engulfment and cell motility protein 1 (ELMO1), key regulators of the phagocytic process, were upregulated following treatment. Complement components CFH and C3 increased substantially, and brain C3 correlated with circulating C3 (Fig. 5m and Extended Data Fig. 11d,e), consistent with systemic-to-CNS innate immune priming. The rise in C3, implicated in microglial Aβ clearance, coincided with reduced cytochrome P450 family 51 (CYP51) and LRP receptor-related protein-associated protein 1 (LRPAP1), two negative regulators of microglia-dependent Aβ removal (Fig. 5m). Notably, ICAM1, FCGR2B and CFH, already elevated in AD\u003csup\u003e+\u003c/sup\u003e mice, were further increased by ETP69, suggesting that H3K9me3 inhibition amplifies an existing neuroimmune response to enhance microglia- and monocyte-mediated chemotaxis and phagocytosis, thereby promoting more efficient cerebral Aβ clearance. \u0026nbsp;Together, these proteomic signatures indicate that lowering cerebral H3K9me3 broadly resets AD brains toward a WT-like state by coordinately restoring synaptic–cytoskeletal programs and potentiating complement-linked phagocytic immunity to support Aβ clearance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRestored neurotrophic VGF–granin signalling tracks cognition\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe next examined whether H3K9me3 inhibition engages molecular programs linked to neuroprotection and behaviour. IPA predicted significant activation of pathways and upstream regulators related to neuroplasticity as well as learning and memory, reversing the deficits observed in AD\u003csup\u003e+\u003c/sup\u003e mice (Fig. 6a-c and Extended Data Fig. 12a–c); a STRING network highlighted shared DEPs across these functions (Extended Data Fig. 12d). Among these, leucine-rich repeat neuronal protein 4 (LRRN4), a regulator of hippocampus-dependent learning and long-lasting memory, was strongly induced in ETP69-treated versus control AD\u003csup\u003e+\u003c/sup\u003e brains (1.73-fold, p = 0.015). Five proteins were common to all three IPA functions—NTRK2 (TRKB), SHANK3, SPAST, TSC2 and NGF-inducible neurosecretory protein (VGF; Extended Data Fig. 12d). Notably, IPA identified RICTOR, GABA and BDNF activation by ETP69 treatment (z scores: 3.59, 2.50, 2.45, respectively), whereas these regulators were among the most strongly inhibited in AD\u003cem\u003e\u003csup\u003e+\u003c/sup\u003e\u003c/em\u003e mice (Fig. 6c). Given the key role of BDNF in neuronal survival, synaptic plasticity, learning, and memory, we further analyzed downstream targets in the BDNF network (Fig. 6d). The prediction of BDNF activation was based on the downregulation of phospholipid transfer protein (PLTP) and annexin A2 (ANXA2), which was further validated by hippocampal analysis of aged AD\u003cem\u003e\u003csup\u003e+\u003c/sup\u003e\u003c/em\u003e mice treated with ETP69 (Fig. 6e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eImportantly, BDNF activation network identified two proteins at the intersection of learning, dendritic spine maintenance, and neuronal survival, VGF and neurotrophic receptor tyrosine kinase 2 (NTRK2; a BDNF receptor; Fig. 6d and Extended Data Fig. 12d). Indeed, ETP69 induced cerebral NTRK2 in both WT and AD\u003csup\u003e+\u003c/sup\u003e mice alongside upregulating the proprotein convertase subtilisin/kexin type 1 (PCSK1) in AD\u003csup\u003e+\u003c/sup\u003e mice (Fig. 6f); the latter is a convertase that cleaves VGF into TLQP-62 and TLQP-21 active neuropeptides. Given VGF’s membership in the granin family, we assessed granin-related proteins and found that VGF, CHGA, CHGB/SCG1, SCG2, SCG5 (7B2) and PCSK1N/SCG8 were reduced in AD\u003csup\u003e+\u003c/sup\u003e versus WT brains and shifted upward to WT-like levels after ETP69; SCG3 showed the opposite pattern and was similarly normalized (Fig. 6g). We further validated BDNF and VGF expression in two independent cohorts using WB and IHC (Fig. 6h−j). H3K9me3 inhibition led to hippocampal BDNF induction (WB: 1.3-fold, p = 0.0284; IHC: 1.3-fold, p = 0.0599), whereas VGF increased robustly in both aged WT (WB: 1.3-fold, p = 0.0017; IHC: 1.4-fold, p = 0.0215) and AD\u003csup\u003e+\u003c/sup\u003e mice (IHC: 1.9-fold, p \u0026lt; 0.0001; WB: 1.4-fold, p = 0.0004), with prominent signal in the dentate gyrus subgranular zone (Fig. 6j). Cortical VGF was reduced in AD\u003csup\u003e+\u003c/sup\u003e versus WT mice (32%, p = 0.0003) and was restored by ETP69 (1.9-fold, p = 0.0075). Alongside decreased H3K9me3 (28%, p = 0.0069), amyloid plaques were decreased in the hippocampus (12F4-Aβ: 27%, p = 0.0015; ThioS: 38%, p = 0.0333) and cortex (ThioS: 17%, p = 0.0080) following ETP treatment (Fig. 6h−j). Importantly, cerebral VGF inversely correlated with H3K9me3 in the hippocampus (IHC: r = −0.64, p = 0.0022; WB: r = −0.60, p = 0.0412) and cortex (r = −0.68, p = 0.0214), and associated with reduced Barnes maze errors reflecting improved memory retention (r = −0.58, p = 0.0088; Fig. 6k), highlighting VGF as a direct mechanistic bridge from epigenetic modulation to cognitive benefit.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRetinal immune–synaptic proteostasis and colour vision rescue by SUV39H1 inhibition\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo explore the impact of H3K9me3 inhibition on the retina of AD\u003csup\u003e+\u003c/sup\u003e mice, we studied the MS-based proteome profiles of ETP69-treated 14-month-old WT and AD\u003csup\u003e+\u003c/sup\u003e cohort mice (Fig. 7 and Extended Data Figs. 13-14). We detected 5,107 proteins in ≥3 animals per group and identified 732 DEPs (p \u0026lt; 0.05) across AD\u003csup\u003e+\u003c/sup\u003e/WT (226), WT ETP69/DMSO (452) and AD\u003csup\u003e+\u003c/sup\u003e ETP69/DMSO (153) comparisons (Fig. 7a). As in brain, WT and AD\u003csup\u003e+\u003c/sup\u003e treatment responses minimally overlapped (11 DEPs), and brain–retina concordance was limited (Extended Data Fig. 13a; only three shared DEPs in either WT ETP69/DMSO or AD\u003csup\u003e+\u003c/sup\u003e ETP69/DMSO), indicating genotype- and tissue-specific drug effects. Heatmaps and PCA segregated samples by genotype and treatment (Fig. 7b and Extended Data Fig. 13b,c). Notably, all 49 DEPs shared between AD\u003csup\u003e+\u003c/sup\u003e/WT and AD\u003csup\u003e+\u003c/sup\u003e ETP69/DMSO were fully reversed by treatment in AD\u003csup\u003e+\u003c/sup\u003e retina (Fig. 7c). Volcano plots highlight the top 20 up- and downregulated DEPs in treated versus control AD\u003csup\u003e+\u003c/sup\u003e retina (Fig. 7d; AD\u003csup\u003e+\u003c/sup\u003e/WT and WT ETP69/DMSO in Extended Data Fig. 14a; DEPs with |FC| \u0026gt; 1.20 in Extended Data Tables 8–13).\u003c/p\u003e\n\u003cp\u003eGO enrichment across 330 DEPs (177 AD\u003csup\u003e+\u003c/sup\u003e/WT, 104 AD\u003csup\u003e+\u003c/sup\u003e ETP69/DMSO and 49 shared) identified AD-driven perturbations, and ETP69 caused restoration, of apoptosis, synapse organization/dendritic spine and neurotransmission, and acute immune response pathways (Fig. 7e,f). Consistent with our brain and blood datasets, HP and HPX were among the top induced retinal DEPs in ETP69-treated AD\u003csup\u003e+\u003c/sup\u003e mice (HP FC = 3.28, p = 0.0084; HPX FC = 2.18, p = 0.0319), with similar but non-significant trends in WT (HP FC = 1.63, p = 0.0997; HPX FC = 1.30, p = 0.1315; Extended Data Table 12). Orosomucoid 2 (ORM2) was likewise strongly upregulated in both genotypes (Fig. 7d and Extended Data Fig. 14a). In AD\u003csup\u003e+\u003c/sup\u003e retina, ETP69 reduced pro-inflammatory coagulation factor III (F3) and thrombin (F2), an IPA-predicted upstream regulator (Fig. 7f and Extended Data Fig. 14b). IPA further predicted activation of TNF and SP1 in AD\u003csup\u003e+\u0026nbsp;\u003c/sup\u003eretina (z = 2.43 and 2.38), with ETP69 fully reversing TNF (z = −2.02) and partially attenuating SP1 (z = −0.97; Fig. 7f and Extended Data Fig. 14b), consistent with modulation of retinal Aβ-, apoptosis- and inflammation-linked signalling.\u003c/p\u003e\n\u003cp\u003eAll DEPs with histone-modifying activity in AD\u003csup\u003e+\u003c/sup\u003e retina were reversed by ETP69 (Fig. 7e). Specifically, RIOX1, PPP6C and the retina-enriched kinase STK35 (reduced in AD\u003csup\u003e+\u003c/sup\u003e retina) were increased by treatment, whereas the H3K9 deacetylase SIRT1 (elevated in AD\u003csup\u003e+\u003c/sup\u003e retina) was decreased. ETP69 also increased RSBN1 (H4K20 demethylase activity) while reducing CLOCK and GSK3A in AD\u003csup\u003e+\u003c/sup\u003e retina. Importantly, IPA of direct/indirect neighbour DEPs predicted SUV39H1 activation in AD\u003csup\u003e+\u003c/sup\u003e retina (Extended Data Fig. 14c) and its inhibition after ETP69 (Fig. 7g); SUV39H1, SIRT1, CLOCK and GSK3A are established modulators of circadian rhythm, ageing and retinal disease.\u003c/p\u003e\n\u003cp\u003eSynapse- and neurotransmission-related pathways comprised a major component of AD\u003csup\u003e+\u003c/sup\u003e retinal dysregulation and were broadly restored by ETP69 (Fig. 7e). PICK1, which regulates AMPAR internalization and synaptic plasticity, was reduced by treatment (and elevated in AD\u003csup\u003e+\u003c/sup\u003e retina), whereas ADGRL3 (postsynaptic spine-enriched) and CRIPT (anchors PSD95 at excitatory synapses) were increased in treated AD\u003csup\u003e+\u003c/sup\u003e retina (Extended Data Fig. 14b), consistent with improved synapse-supportive programs.\u003c/p\u003e\n\u003cp\u003eIn parallel to the brain, ETP69 also modulated extended granin family members in AD\u003csup\u003e+\u0026nbsp;\u003c/sup\u003eretina: CHGA and PCSK1N were upregulated in treated AD\u003csup\u003e+\u003c/sup\u003e retina (with PCSK1N also increased in WT), whereas VGF was uniquely reversed and downregulated by ETP69 in AD\u003csup\u003e+\u003c/sup\u003e retina—opposite to brain (Fig. 7h), consistent with reports of elevated VGF in retinal neurodegeneration and its prominent expression in Müller/astrocytic glia. GO terms further implicated visual behaviour and learning (Fig. 7e). In colour-mode X-maze testing, WT alternations preferentially followed continuous W↔R↔G↔B sequences, a pattern reduced in AD\u003csup\u003e+\u003c/sup\u003e mice but restored by ETP69 (Fig. 7i); these sequences typically began or ended at the white arm, efficiently avoiding bidirectional B↔W transitions (Fig. 7i). Together, these data show that H3K9me3 inhibition selectively reprograms the AD\u003csup\u003e+\u003c/sup\u003e retinal proteome toward WT-like synaptic and immune homeostasis, restoring visual learning.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we identify the repressive heterochromatin mark H3K9me3 as an early, cross-compartment epigenetic signature of AD across the brain\u0026ndash;retina axis, consistent with a shift toward chromatin compaction that constrains neuronal and immune programs. In matched human donor brain and retinal tissues, H3K9me3 was already elevated at the MCI stage, tracked neuropathological burden and cognitive dysfunction, and showed strong brain\u0026ndash;retina concordance. Proteomic profiling of AD cortex and retina indicated enrichment of pathways governing heterochromatin formation and histone modifications, predicting SUV39H1 activation and H3K9 hypertrimethylation. Mechanistically, our interventional studies in transgenic AD models revealed that pharmacologic inhibition of SUV39H1 rapidly and durably lowers cerebral H3K9me3, recalibrates brain and retina neuroimmune competence and synaptic proteostasis, attenuates gliosis and amyloid/tau pathology, culminating in rescue of dendritic spine architecture and visuo-cognitive function. These benefits are attributed, at least in part, to a tissue-specific restoration of a trophic\u0026ndash;immune axis, centered on neurosecretory VGF and the extended granin family, bridging epigenetic remodeling with functional recovery. Together, these observations support a model in which age- and disease-associated epigenomic regulation at the brain\u0026ndash;retina interface can actively stabilize maladaptive cellular states rather than simply reflect downstream pathology.\u003c/p\u003e\n\u003cp\u003eThe synchronized H3K9me3 abundance between the dorsolateral prefrontal cortex and the superotemporal retina across cognitively normal individuals and patients spanning the AD continuum positions the retina as an accessible CNS readout of epigenetic mark of disease severity. In both compartments, H3K9me3 rises from the MCI stage and further in AD, tightly correlating with amyloid and tau proteinopathy, and tracking disease stage (ABC, Braak) and cognitive impairment (CDR, MMSE). These observations are consistent with prior reports of elevated H3K9me3 in the hippocampus and in orbitofrontal and temporal cortices of individuals with AD\u003csup\u003e39,62,63\u003c/sup\u003e. The direct correlation between retinal and prefrontal cortical H3K9me3 is in line with\u0026nbsp;mounting evidence that the retina recapitulates key molecular and pathological features of AD\u003csup\u003e13,14,16-20,25,27,64-68\u003c/sup\u003e. Proteomics in human brain and retina tissues converge on chromatin-related dysregulation consistent with enhanced H3K9me3, yet tissue-specific: in the brain, altered PRMT1/PRMT5 expression and prediction of SUV39H1 activation supported increased H3K9me3, while in the retina, downregulation of KDM4B/C and KAT7 together with increased JMJD6 indicated the same shift. Although the specific molecular players differ between tissues, consistent with their distinct cellular compositions, the directionality of epigenetic remodelling is preserved, indicating a conserved disease mechanism rather than coincidental tissue-specific changes. Future studies should delineate which epigenetic biomarkers are shared across compartments and which reflect tissue-specific regulatory programs in healthy aging and disease.\u003c/p\u003e\n\u003cp\u003eBuilding on these human data, we demonstrate a causal contribution of excessive H3K9me3 in AD-like phenotypes and that targeting H3K9me3-dependent repression is disease modifying in preclinical AD models. Our studies provide the first evidence that pharmacological SUV39H1-H3K9me3 inhibition (ETP69) in midlife and aged AD mouse models (APPswe/PS1dE9 and 3xTg-AD) attenuates amyloidosis and tau pathology, dampens neuroinflammation, and restores synaptic integrity and cognitive performance. Prior work also showed the effect of ETP69 in aged WT mice\u003csup\u003e69\u003c/sup\u003e. In our study, lowering H3K9me3 in AD-model mice reactivated cerebral and retinal molecular pathways governing synaptic plasticity, dendritic spine dynamics, neuronal survival, and learning, while simultaneously reshaping the neuroimmune environment toward enhanced amyloid clearance and reduced gliosis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDendritic spine remodelling is central to learning and memory\u003csup\u003e70,71\u003c/sup\u003e, and our proteomics data predicted increases in spine density and branching by H3K9me3 inhibition. Structural analyses by Golgi-Cox confirmed the emergence\u0026nbsp;of new\u0026nbsp;filopodia and long-thin\u0026nbsp;spines\u0026mdash;critical for encoding new memories\u003csup\u003e72\u003c/sup\u003e\u0026mdash;in ETP69-treated AD-model mice to levels comparable to or exceeding WT, providing a morphological correlate for cognitive recovery. At the molecular level, IPA predicted activation of the BDNF network, with marked upregulation of BDNF receptor NTRK2 in AD\u003csup\u003e+\u003c/sup\u003e mice, exceeding WT levels, following treatment. Given that \u003cem\u003eBdnf\u003c/em\u003e transcription is epigenetically regulated\u003csup\u003e73\u003c/sup\u003e and sensitive to H3K9 di- and tri-methylation\u003csup\u003e69,74,75\u003c/sup\u003e, and that CREB\u0026ndash;BDNF\u0026ndash;NTRK2 signalling is diminished in AD\u003csup\u003e76-79\u003c/sup\u003e, restoration of this pathway provides mechanistic link to cognitive rescue.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA key BDNF effector\u003csup\u003e80\u003c/sup\u003e, VGF was increased in brains of ETP69-treated AD\u003csup\u003e+\u003c/sup\u003e mice to\u0026nbsp;levels equal or higher to that of WT mice, and inversely correlated with H3K9me3 levels. Harmonizome data identify \u003cem\u003eVgf\u0026nbsp;\u003c/em\u003eamong the 401 genes with H3K9me3-rich promoters\u003csup\u003e81\u003c/sup\u003e, suggesting direct epigenetic derepression of Vgf upon H3K9me3 reduction. The protease PCSK1 that cleaves VGF into bioactive peptides TLQP-62 and TLQP-21\u003csup\u003e82\u003c/sup\u003e, was upregulated alongside VGF. TLQP-62 has been implicated in dendritic branching, synaptic plasticity and long-term memory\u003csup\u003e80,83-85\u003c/sup\u003e, and its reported roles in neurogenesis\u003csup\u003e86,87\u003c/sup\u003e are consistent with the prominent VGF induction we observed in the dentate gyrus subgranular zone. Since TLQP-62 activates VGF translation via mTOR signalling\u003csup\u003e88\u003c/sup\u003e and requires NTRK2 for hippocampal memory formation\u003csup\u003e80\u003c/sup\u003e,\u0026nbsp;our findings support an mTOR-mediated, positive PCSK1\u0026ndash;VGF\u0026ndash;NTRK2 regulatory loop that reinforces the cognitive benefits of H3K9me3 inhibition.\u003c/p\u003e\n\u003cp\u003eOther members of the extended granin family\u003csup\u003e89\u003c/sup\u003e, including CHGA, CHGB, SCG2, 7B2, and PCSK1N, were upregulated following treatment. Multiple omics studies have shown downregulation of cerebral VGF along with, albeit to a lesser extent, all other granin family members (CHGA, CHGB, SCG2, SCG3, 7B2 and PCSK1N) in AD patients\u003csup\u003e90-104\u003c/sup\u003e as well as in a murine AD model\u003csup\u003e91,100\u003c/sup\u003e. As granins regulate vesicles biogenesis and neuropeptide release critical for neuronal activity and cognitive function\u003csup\u003e105-108\u003c/sup\u003e, their restoration likely contributes to improved synaptic transmission.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBeyond synaptic function, granins and VGF-derived peptides intersect with innate immunity and A\u0026beta; processing and clearance. Granin family members localize near A\u0026beta; plaques in brains of AD patients and mouse models, exerting neuroprotective effect; 7B2 and PCSK1N inhibit amyloid aggregation\u003csup\u003e121,122\u003c/sup\u003e and efficiently prevent A\u0026beta; neurotoxicity. VGF-derived peptide TLQP-21 enhances microglial phagocytosis and chemotaxis via C3AR1 to reduce A\u0026beta; plaque load and dystrophic neurites\u003csup\u003e123-125\u003c/sup\u003e, consistent with reduced cerebral A\u0026beta; burden and a dystrophic neurite marker LAMP1 in ETP69-treated AD-model mice. Treatment further increased circulating and brain-infiltrating monocytes and elevated innate immune mediators, including haptoglobin, hemopexin and complement C3; in parallel, H3K9me3 targeting increased cerebral ICAM1 with its receptor ITGAM/ITGB2\u003csup\u003e109\u003c/sup\u003e, which also binds C3\u003csup\u003e110\u003c/sup\u003e, suggesting augmented monocyte recruitment and phagocytic capacity, in line with evidence that peripheral monocyte enrichment promotes A\u0026beta; clearance and preserves cognition\u003csup\u003e54-61,111-120\u003c/sup\u003e. Collectively, these data indicate that H3K9me3 acts as a trophic\u0026ndash;immune checkpoint inadequately engaged in AD, and its inhibition enhances the brain\u0026rsquo;s capacity to counter amyloidosis and chronic neuroinflammation.\u003c/p\u003e\n\u003cp\u003eIn the retina, ETP69 robustly activated synaptic and immune programs, yet treatment-responsive DEPs showed minimal overlap with the brain, consistent with tissue-specific proteomic networks. VGF regulation was notably divergent, increased in AD\u003csup\u003e+\u003c/sup\u003e retina but decreased in AD\u003csup\u003e+\u003c/sup\u003e brain, yet in both compartments it shifted toward WT-like levels with treatment, highlighting the decisive influence of local cell-type context on epigenetic-to-proteomic outputs; in the retina, VGF is predominantly expressed by M\u0026uuml;ller glia and astrocytes, whereas in the brain it is primarily neuronal. Like the brain, retinal granins including CHGA and PCSK1N were elevated by ETP69. Although acute-phase proteins HP and HPX were induced across brain, retina and blood, treatment elicited a retina-selective, disease-corrective proteomic response that opposed AD-associated apoptosis and TNF immune activation leading to rescue of synapse and colour vision.\u003c/p\u003e\n\u003cp\u003eOur study integrates cross-compartment human datasets with mechanistic, interventional validation in midlife and aged AD mouse models, yet several considerations remain. First, the human analyses are cross-sectional and correlative, limited by\u0026nbsp;sample size and heterogeneity; larger, longitudinal studies will be needed to determine the temporal dynamics of H3K9me3 and its relationship to disease onset and progression in humans.\u0026nbsp;Second, the reproducible molecular and functional rescue elicited by SUV39H1\u0026ndash;H3K9me3 inhibition across AD models motivates deeper mechanistic resolution, including cell-type\u0026ndash;resolved epigenomic interrogation using chromatin accessibility and single-cell transcriptomic approaches. Third, translation will require safety and efficacy testing of\u0026nbsp;H3K9me3-targeted\u0026nbsp;treatment in AD patients, alongside development and clinical validation of noninvasive retinal readouts for specific chromatin marks to establish the retina as a proxy for cerebral epigenetic state.\u003c/p\u003e\n\u003cp\u003eIn conclusion, this study defines H3K9 hypertrimethylation as an\u0026nbsp;epigenetic synaptic\u0026ndash;immune checkpoint\u0026nbsp;linked to neurodegeneration in AD.\u0026nbsp;This repressive heterochromatin signature spans brain and retina in\u0026nbsp;human AD and preclinical models. Pharmacologic SUV39H1\u0026ndash;H3K9me3\u0026nbsp;inhibition reconfigures chromatin to restore synaptic plasticity and immune competence programs, culminating in recovery of cognitive and visual function. By coordinately resetting multiple dysregulated pathways, epigenomic modulation emerges as a transformative alternative to therapies that solely target proteotoxic clearance. These findings resonate with growing interest in epigenetic reprogramming to counteract ageing and neurodegeneration, including ocular ageing where epigenetic drift contributes to visual decline\u003csup\u003e126,127\u003c/sup\u003e. The brain\u0026ndash;retina concordance further positions the retina as a noninvasive surrogate for epigenetic biomarker imaging to guide early precision diagnosis and therapy.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eMice\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDouble-transgenic B6.Cg-Tg (APPswe/PSEN1dE9) 85Dbo/Mmjax mice, a mouse model of Alzheimer\u0026rsquo;s disease (AD\u003csup\u003e+\u003c/sup\u003e), and their age-matched WT C57BL/6J littermates were obtained from the Mutant Mouse Resource and Research Center (MMRRC) at the Jackson Laboratory (RRID:MMRRC_034832-JAX). Mice were bred and maintained at Cedars-Sinai Medical Center vivarium under standardized conditions: housed up to five per cage on a 14-hour light/10-hour dark cycle, ambient temperature maintained at 74\u0026thinsp;\u0026deg;F (23\u0026thinsp;\u0026deg;C) \u0026plusmn;\u0026thinsp;2\u0026thinsp;\u0026deg;F, and relative humidity at 30\u0026ndash;70%, with \u003cem\u003ead libitum\u003c/em\u003e access to food and water and a maximum of five animals per cage. All AD\u003csup\u003e+\u003c/sup\u003e animals used in this study had a congenic C57BL/6 background. Both male and female mice were used for all experiments and were assigned to experimental groups after balancing for age and genotype. A cohort of female triple-transgenic B6;129-Tg (APPswe/tauP301L)1Lfa \u003cem\u003ePsen1\u003csup\u003etm1Mpm\u003c/sup\u003e\u003c/em\u003e (3xTg AD\u003csup\u003e+\u003c/sup\u003e) mice (Jackson Laboratories, MMRRC_034830-JAX), maintained at the University of California, Irvine (UCI), was included for a limited subset of experiments and analysed separately where indicated.\u003c/p\u003e\n\u003cp\u003eAll procedures\u0026nbsp;were approved by the Institutional\u0026nbsp;Animal Care and Use Committee (IACUC) of Cedars-Sinai Medical Center and UCI and conducted in accordance with the NIH and ARRIVE Guidelines for the Care and Use of Laboratory Animals.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eETP69 treatment regimen\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLyophilized ETP69 was resuspended in 50% DMSO (in saline) at a concentration of 10 mg/ml.\u003c/p\u003e\n\u003cp\u003eETP69-treated mice received intraperitoneal injections of ETP69 (10 mg/kg), and control mice (DMSO-injected groups) received vehicle injections (50% DMSO in saline). Four cohorts of WT and AD\u003csup\u003e+\u003c/sup\u003e mice (n=153; 14 and 18 months old) were treated with ETP69 or DMSO under three regimens (regimens S, R, and B), with behavioural timelines and experimental endpoints shown in Fig. 2b (18-month-old cohort) and Fig. 4d (14-month-old cohort). Two cohorts of AD\u003csup\u003e+\u003c/sup\u003e and WT mice received a single ETP69 dose (regimen S) 1 day before the start of behavioural testing (day 0) and were euthanized either on day 4 or day 15. A third group of AD\u003csup\u003e+\u003c/sup\u003e and WT mice received repeated ETP69 doses (once weekly for 11 weeks) with the last administration on day 0 (regimen R). A fourth group of AD\u003csup\u003e+\u003c/sup\u003e and WT mice received the first dose of ETP69 on day 0 and a booster dose on day 9, the day before memory retention testing of the Barnes maze (regimen B). Control AD\u003csup\u003e+\u003c/sup\u003e and WT mice received DMSO according to the same regimens S, R, or B. One cohort of female 3xTg AD\u003csup\u003e+\u003c/sup\u003e mice (n=11; 14 months old) received a single injection of ETP69 or DMSO (Fig. 4a).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBehavioural tests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll WT and AD\u003csup\u003e+\u003c/sup\u003e animals in this study underwent behavioural tests.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOpen field test\u003c/p\u003e\n\u003cp\u003eSpontaneous locomotor activity was assessed for 30 min using the Photobeam Activity System (www.sandiegoinstruments.com). Ambulatory and rearing activities were recorded, and distance, speed, and resting time were calculated.\u003c/p\u003e\n\u003cp\u003eBarnes maze test (hippocampus-based spatial learning and memory test)\u003c/p\u003e\n\u003cp\u003eMice were first trained to locate an escape box in a 20-hole circular table during a 4-min trial that was performed 3 times per day for 4 days (acquisition training phase), as previously described \u003csup\u003e57\u003c/sup\u003e. Following a 2-day break, the memory retention of each mouse was evaluated on day 7 (retention phase). Memory extinction and learning of a new escape location were assessed on days 8\u0026ndash;9 (reversal phase). The latency to find the escape box and the number of incorrect entries (errors) were recorded for each trial and averaged on each day for each mouse.\u003c/p\u003e\n\u003cp\u003eY-maze spontaneous alternation test (hippocampus-based spatial working memory test)\u003c/p\u003e\n\u003cp\u003eThe Y-shaped apparatus used for this study consists of two arms equal in length and one longer arm. Mice\u0026nbsp;were individually placed at the distal end of the long arm and allowed to move freely throughout the entire maze (all three arms) for 5 min under dim light. The sequence of arm entries and the total number of entries were recorded; the percentage of spontaneous alternations was calculated as follows:\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" width=\"609\" height=\"64\"\u003e\u003c/p\u003e\n\u003cp\u003eA spontaneous alternation is defined as the sequential visit to the three different arms without returning to a previously visited one.\u003c/p\u003e\n\u003cp\u003eVisual-stimuli X-maze test (visual-cognitive test)\u003c/p\u003e\n\u003cp\u003eTo assess spontaneous behaviour induced by colour (under equal conditions) and contrast sensitivity, the mice were tested using our custom-made colour and contrast-mode X-maze, as previously described \u003csup\u003e27,128,129\u003c/sup\u003e. For each mode, the mice were individually placed in the centre of the ViS4M and allowed to freely explore the maze for 5 min. The sequences of arm entries were manually documented according to the video recordings. The total number of entries, percentage of bidirectional transitions between arms, and percentage of alternations were all determined from the sequences of arm entries.\u003c/p\u003e\n\u003cp\u003eChord diagrams were generated to visualize behavioural data from the Barnes maze and X-maze tests using the free online resource Circos (mkweb.bcgsc.ca/tableviewer/) as previously described \u003csup\u003e128,129\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eContext-specific fear conditioning test (spatial associative memory test)\u003c/p\u003e\n\u003cp\u003eDuring the acquisition phase, an animal was placed in a freezing behaviour-monitoring chamber (www.sandiegoinstruments.com) and allowed to habituate for 2 min before receiving a 0.2-mA electric foot shock for 1 s. The animal remained in the chamber for an additional 3 min and was then returned to its home cage. To assess context-specific fear, the animal was placed in the same chamber 24 h after the acquisition phase, and the freezing time (absence of movement for 3 s) over a 4-min session was recorded.\u003c/p\u003e\n\u003cp\u003eAll behavioural tests were performed by an experimenter blinded to mouse genotypes and treatments.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMouse brain collection and processing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAfter completion of the behavioural tests, all mice were deeply anaesthetized (50 mg/kg ketamine/xylazine) and transcardially perfused with ice-cold saline solution containing 0.5 mM EDTA. Mouse brains were collected and processed as follows: 1) snap frozen and stored at \u0026minus;80\u0026deg;C for protein extraction; 2) fixed in 2.5% PFA overnight and then cryoprotected in 30% sucrose for immunohistological analyses; or 3) processed for Golgi-Cox staining. Fixed brains were coronally sectioned at 30 \u0026micro;m thickness using a cryostat (Leica CM3050 S; Leica Biosystems, Nussloch, Germany). Sections were stored at 4\u0026deg;C in PBS containing 0.01% sodium azide in 24-well plates, until immunochemical processing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePostmortem human brains and retinas\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHuman brain and retinal tissues were obtained from the Alzheimer\u0026rsquo;s Disease Research Center (ADRC) Neuropathology Core in the Department of Pathology (IRB protocol HS-042071) of Keck School of Medicine at the University of Southern California (USC, Los Angeles). USC-ADRC maintains human tissue collection protocols that are approved by institutional managerial committees and subject to oversight by the National Institutes of Health. The ADRC provided clinical and neuropathological reports on patients\u0026rsquo; neurological examinations, neuropsychological and cognitive tests, family history, and medication lists, as collected in the ADRC system using the Uniform Data Set (UDS).\u0026nbsp;Histological studies were performed under an approved IRB protocol at Cedars-Sinai Medical Center.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe examined brains and retinas from deceased patient donors with clinically and neuropathologically confirmed AD (n=13) or mild cognitive impairment (MCI due to AD, n=6) as well as brains from deceased individuals with normal cognition (CN [control], n=6). Donor information is provided in Extended Data Table 1. Fresh brain tissues (frontal cortex) were snap frozen and stored at \u0026minus;80\u0026deg;C. Portions of fresh-frozen brain tissues were fixed in 4% PFA for 16 h and then dehydrated in 30% sucrose/PBS. The brain tissues were coronally sectioned (30 \u0026mu;m thick) on a cryostat (Leica CM 3050_S) and mounted on slides coated with 3-aminopropyltriethoxysilane (#A3648 Sigma-Aldrich). The sections were then treated with target retrieval solution (pH 6.1; S1699, Dako) at 99\u0026deg;C for 40 min and washed with PBS before being used for immunohistochemistry (IHC).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe processing of eye globes, isolation and preparation of retinal strips, and retinal immunostaining were extensively detailed in \u003csup\u003e14,16,19,27\u003c/sup\u003e. Briefly, donor eyes were collected within an average of 9 hours after death, puncture at the limbus and fixed in 10% neutral buffered formalin (NBF) or 4% paraformaldehyde (PFA) then stored at 4\u0026deg;C. Fixed eyes were dissected as previously described (same refs as above). Flatmounts were prepared after careful dissection of the eye globes and thorough cleaning of the vitreous humor. Flatmount strips (~2 mm wide) extending diagonally from the optic disc (OD) to the ora serrata (~20\u0026ndash;25 mm long) were prepared in 4 predefined regions: Superior Temporal (ST), Inferior Temporal (IT), Inferior Nasal (IN), and Superior Nasal (SN). In this study, we focused our analysis on the ST retinal strip due to the high presence of AD pathology in this region. The flatmount-derived strips were then paraffinized using standard techniques and embedded in paraffin after flip-rotating 90\u0026deg; horizontally. The retinal strips were sectioned (7-10 \u0026micro;m thick) and mounted on microscope slides coated with 3-aminopropyltriethoxysilane.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImmunohistochemical analysis\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMouse or human brain sections mounted on slides were treated with serum-free protein blocking solution (X0909, Dako), then incubated overnight at 4\u0026deg;C with the primary antibodies (sources and dilutions in Extended Data Table 14). Sections were subsequently incubated for 1 h at room temperature with host-specific fluorophore-conjugated secondary antibodies and coverslipped using ProLong Gold Antifade Mountant with DAPI (Molecular Probes, Life Technologies). In some cases, slides were dipped in Thio-S solution for 1 min (to stain mature\u0026nbsp;A\u0026beta;\u0026nbsp;plaques) after the secondary antibody step and then washed in three baths of 70% ethanol (1 min each) before mounting with ProLong Gold DAPI. Negative controls were processed using the same protocol but without a primary antibody to assess nonspecific labelling.\u003c/p\u003e\n\u003cp\u003eRetinal sections were deparaffinized using 100% xylene twice (10 minutes each), rehydrated with decreasing concentrations of ethanol (100% to 70%), and washed with distilled water followed by PBS. After deparaffinization, tissue sections were treated with target antigen retrieval solution (pH 6.1; S1699, Dako) at 98\u0026deg;C for 1 hour and then washed with PBS. Following steps included blocking, primary and secondary antibodies incubations as described above.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGolgi-Cox staining\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMouse brain hemispheres were processed using a Hito Golgi-Cox OptimStainTM Kit (#HTKNS1125, Hito) according to the manufacturer\u0026rsquo;s instructions. Briefly, brain samples were immersed in impregnation solution in the dark at room temperature for 2 weeks and then cryoprotected at 4\u0026deg;C.Next, the tissues were embedded in OCT compound, sectioned at a thickness of 100\u0026minus;200 \u0026mu;m, mounted on gelatine-coated slides (#HTHS0102, Hito) and allowed to dry overnight in a dark room.The sections were then incubated in a 20% ammonia solution, dehydrated in ascending concentrations of ethanol (50%, 75%, 95%, and 100%), cleared in xylene, and mounted with Permount mounting medium (#SP15-500, Fisher Scientific).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMicroscopy and quantification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eImmunofluorescence images of human and mouse tissue sections were repeatedly captured at 20\u0026thinsp;\u0026times; (resolution of 1388 \u0026times; 1040 pixels, 447.63 \u0026micro;m \u0026times; 335.30 \u0026micro;m /per image) or 40\u0026thinsp;\u0026times; (resolution of 1388 \u0026times; 1040 pixels, 223.82 \u0026micro;m \u0026times; 167.70 \u0026micro;m /per image) at the same focal planes with the same exposure time for each marker. Images were randomly acquired as follows: 3 from the central, 4 from the mid-, and 3 from the far-retinal subregions per human retinal strip (one strip per donor); 5\u0026ndash;7 images per human brain section (one section per donor); and 5\u0026ndash;12 images per mouse brain section (2\u0026ndash;3 sections per animal). Images were exported to Fiji ImageJ (version 2.14.0) to analyse parameters of interest. Acquired images were converted to grayscale and standardized to baseline by using a histogram-based threshold in Fiji ImageJ. This baseline-derived threshold was then applied uniformly to the corresponding single channel for all subjects across diagnostic groups. Images were subsequently subjected to ImageJ2/Fiji particle analysis for each biomarker to determine total and % immunoreactive area.\u003c/p\u003e\n\u003cp\u003eTo assess levels of H3K9me3 in neuronal versus nonneuronal cells, the intensity of H3K9me3 IR in the nuclei (manually delimited with the polygon tool of ImageJ) of NeuN\u003csup\u003e+\u003c/sup\u003e (average of 30 nuclei/animal), GFAP\u003csup\u003e+\u003c/sup\u003e (average of 10 nuclei/animal), or IBA1\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e (average of 20 nuclei/animal) cells was measured at high magnification. H3K9me3 IR in specific regions of the hippocampus (the cornu ammonis [CA] and dentate gyrus [DG] regions) was quantified within areas manually delimited with the polygon tool of ImageJ or Fiji.\u003c/p\u003e\n\u003cp\u003eIn Golgi-Cox\u0026ndash;stained brain sections (Fig. 3b,c and Extended Data Fig. 6a), dendrite segments of pyramidal neurons in the cerebral cortex and the CA1 region of the hippocampus were imaged using a Zeiss ApoTome microscope set at 63\u0026times;\u0026nbsp;(resolution of 1388 \u0026times; 1040 pixels, 225.56 \u0026micro;m \u0026times; 169.01 \u0026micro;m /per image). Dendritic spines were classified as thin (filopodia-like and long-thin), mushroom, or stubby types based on established morphological criteria\u003csup\u003e130\u003c/sup\u003e(Fig. 3d), and manually counted. Image capture and quantification analysis were performed by different investigators.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWestern Blot Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSnap-frozen mouse brain tissues were homogenized in radioimmunoprecipitation assay (RIPA) buffer (0.5M Tris-HCl, pH 7.4, 1.5M NaCl, 2.5% deoxycholic acid, 10% NP-40, 10mM EDTA, Millipore; 20\u0026ndash;188) supplemented with protease and phosphatase inhibitors. Protein concentration was determined using a bicinchoninic acid protein assay kit (Pierce\u003cem\u003eTM\u003c/em\u003e). Lysates were cleared with brief centrifugation for 10 min at 8000g, normalised, and boiled at 95\u0026deg;C after addition of 6X SDS loading dye. Equal amounts of protein (30 \u0026micro;g per sample) were separated on 4\u0026ndash;20% precast polyacrylamide gels (Bio-Rad, catalog #4561094) and transferred to polyvinylidene difluoride membranes. The membranes were then blocked with 2.5% bovine serum albumin in 1x TBS-T (Tris-buffered saline with 0.1% Tween-20) for 1 h at RT, followed by overnight incubation at 4 \u0026deg;C with the primary antibody (sources and dilutions in Extended Data Table 14). After washing with 1x TBS-T, the membrane was incubated with species-specific horseradish peroxidase (HRP)\u0026minus;conjugated or fluorescently labelled secondary antibodies (1:10000). For HRP, the proteins were visualized by incubation with a chemiluminescence substrate kit (#34580, Thermo Fisher). Bands were detected using the LI-COR Odyssey imaging system and quantified using Image Studio software (LI-COR). Relative protein expression levels were calculated by normalizing target protein signals to \u0026beta;-actin or Gapdh. In some cases, membranes were re-probed with primary antibodies after stripping with 1X stripping buffer (NewBlot\u003cem\u003eTM\u003c/em\u003e Nitro stripping buffer, Licorbio #928-40030).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eComplete Blood Count and Hemopexin, Haptoglobin and C3 complement levels\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBlood was collected from the vena cava while the animals were in a state of deep anaesthesia, just prior to transcardiac perfusion, and transferred to EDTA tubes (BD microtainer #365974). Blood was then processed and analysed with the aid of a hematology analyser according to manufacturer instructions (Horiba ABX Micros 60\u0026reg;). Plasma levels of hemopexin, haptoglobin and C3 complement were assessed using ELISA kits # ab157716, ab157714 and ab157711, respectively, according to the manufacturer\u0026rsquo;s instructions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMass spectrometry\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMass spectrometry analysis of human and mouse retina and brain tissues, previously reported in \u003csup\u003e16,55\u003c/sup\u003e, followed these steps: (1) preparation of retinal and brain homogenates; (2) tandem mass tag (TMT) labelling; (3) nanoflow liquid chromatography electrospray ionization tandem mass spectrometry; and (4) database searching, peptide quantification, and statistical analysis.\u003c/p\u003e\n\u003cp\u003eGiven the exploratory nature and limited sample size of our human and mouse datasets, the following criteria were applied to identify significantly differentially regulated proteins (DEPs) for downstream interpretation: 1) at least three measurements per group; 2) an unadjusted P \u0026lt; 0.05; and 3) |FC| \u0026gt; 1.2.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunctional network and computational analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHeatmaps of detectable protein hierarchies were created, and principal component analysis (PCA) were performed using ClustVis (https://biit.cs.ut.ee/clustvis/) \u003csup\u003e131\u003c/sup\u003e. Volcano plots were prepared using\u0026nbsp;GraphPad Prism 10.6.1. software. Pie charts of protein classifications were created using\u0026nbsp;PANTHER (http://pantherdb.org). Gene Ontology (GO, including Biological Process, Molecular Function and Cellular Component) enrichment analysis were performed using MouseMine (www.mousemine.org) and Metascape (https://metascape.org) databases. Pathway networks created in Metascape were subsequently loaded and modified in Cytoscape 3.10.2 (https://cytoscape.org). DEPs were incorporated into molecular pathway and upstream regulator analyses using Ingenuity Pathway Analysis (IPA, Qiagen; https://digitalinsights.qiagen.com). Enrichment and pathway analysis results are reported with z-scores and Benjamini-Hochberg adjusted p-values to control the FDR. Protein interaction networks were generated in String v12.0 (https://string-db.org) and modified in Cytoscape.\u0026nbsp;Volcano plots representing expression changes [log\u003csub\u003e2\u003c/sub\u003e(FC)] and significance level [-log\u003csub\u003e10\u003c/sub\u003e(\u003cem\u003ep\u003c/em\u003e)] were created using GraphPad Prism v10.6.1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistics and reproducibility\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStatistical analyses were performed using GraphPad Prism v10.6.1. Data was analysed using one- or two-way analysis of variance (ANOVA), followed by Fisher\u0026rsquo;s least significant difference (LSD) post hoc test for multiple comparisons (mouse data comparisons limited to: DMSO-AD\u003csup\u003e+\u003c/sup\u003e versus DMSO-WT, ETP69-AD\u003csup\u003e+\u003c/sup\u003e versus DMSO-AD\u003csup\u003e+\u003c/sup\u003e and ETP69-WT versus DMSO-WT). Comparisons between two groups were performed using unpaired, two-tailed Student\u0026rsquo;s t-tests. Correlations were assessed using Pearson\u0026rsquo;s correlation analysis; Pearson\u0026rsquo;s coefficient (r) indicates the strength and direction of linear associations. Data is presented as mean\u0026thinsp;\u0026plusmn;\u0026nbsp;standard error of the mean (SEM) unless otherwise stated. Violin plots display the median and the lower and upper quartiles. Statistical significance was defined as\u0026nbsp;*p \u0026lt; 0.05, **p \u0026lt; 0.01, ***p \u0026lt; 0.001, and ****p \u0026lt; 0.0001.\u003c/p\u003e\n\u003cp\u003eNo statistical methods were used to predetermine sample size. Sample sizes were selected based on prior experience and consistency with previously published studies in the field. No randomization was performed. Data points identified as outliers (2\u0026thinsp;standard deviations away from the mean) were excluded from analysis prior to statistical testing. Data distribution was assumed to be approximately normal but was not formally tested.\u003c/p\u003e\n\u003cp\u003eHistological and western blot analyses were performed once per experiment and included three or more independent biological replicates. Different investigators performed experiments and analyses at different stages of the study. Investigators were blinded to mouse genotype and treatment during behavioral assays. No randomization was performed. Human and animal experimental groups were balanced for age and sex where possible.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMost data generated and analysed in this study are included in this manuscript and extended data. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifiers PXD040225 (human data) and PXD041527 (mouse data).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Institutes of Health (NIH)/National Institute on Aging (NIA) grants R01AG056478, R01AG055865, and AG056478-04S1 (M.K.H.). The work was also supported by the Hertz Innovation Fund, The Saban, Snyder, Wilstein, and The Gordon Private Foundations, and The Jona Goldrich Center Alzheimer\u0026rsquo;s Disease (M.K.H.). The Ray Charles Scholar Foundation supported M.R.D. and J.W.W. The authors thank the\u0026nbsp;Cedars-Sinai Biobehavioral Research Core for assistance with and access to equipment for testing. We thank Drs. Carol Ann Miller and Debra Hawes (ADRC Neuropathology Core; University of Southern California), as well as Drs. Rodrigo Medeiros and Joao A. Paulo (University of California-Irvine and Harvard Medical School, respectively) for providing human brain tissues, analysis, and neuropathological reports. We thank Dr. Min Lin (Horne\u0026rsquo;s lab) for providing ETP69 compound and Dr. Rakez Kayed for previously sharing the T22 oligo-tau antibody. We also thank Samuel Fuchs, Ella Maru Studio, and Biorender.com for illustrations and figure artwork.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStudy conception and design: MKH, DTF, and KLB\u003c/p\u003e\n\u003cp\u003eLive animal experiments: DTF, JPV, JS, YK, OC, and TS\u003c/p\u003e\n\u003cp\u003eHuman and mouse tissue collection, isolation, and processing: DTF, AR, JPV, YK, BPG, JS, OC, HS, SS, and MKH\u003c/p\u003e\n\u003cp\u003eData acquisition, curation, and analysis:\u0026nbsp;DTF, JPV, AR, YK, JS, BPG, SS, HS, MRD, JWW, LSS, and MKH\u003c/p\u003e\n\u003cp\u003eMass spectrometry experiments and analysis: MM, JPV, JS, DTF, YK, SLG,\u0026nbsp;VKG, and MKH\u003c/p\u003e\n\u003cp\u003eStatistical analysis: JPV, DTF, AR, and MKH\u003c/p\u003e\n\u003cp\u003eInterpretation of the data: JPV, DTF, AR, YK, JS,\u0026nbsp;MF, KLB, and MKH\u003c/p\u003e\n\u003cp\u003eExternal resources: DH, MTK, KLB\u003c/p\u003e\n\u003cp\u003eWriting-original draft: JPV, DTF and MKH\u003c/p\u003e\n\u003cp\u003eWriting-editing, review: DTF, JPV, AR, YK, JS, BPG, SS, HS, OC, MRD, JWW, SLG, VKG, LS, MTK, TS, MF, DH, MM, KLB, and MKH\u003c/p\u003e\n\u003cp\u003eStudy supervision: MKH\u003c/p\u003e\n\u003cp\u003eAll authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKLB is the Co-Chairman and shareholder, and MKH is a scientific advisor, of Fortem Neurosciences, Inc. Unrelated to this study: YK, KLB and MKH are co-founders of NeuroVision Imaging, Inc. The other authors have no conflicts to disclose.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eFraga, M. F. \u0026amp; Esteller, M. Epigenetics and aging: the targets and the marks. \u003cem\u003eTrends Genet\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 413-418 (2007). https://doi.org:10.1016/j.tig.2007.05.008\u003c/li\u003e\n\u003cli\u003eBerson, A., Nativio, R., Berger, S. L. \u0026amp; Bonini, N. M. Epigenetic Regulation in Neurodegenerative Diseases. \u003cem\u003eTrends Neurosci\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 587-598 (2018). https://doi.org:10.1016/j.tins.2018.05.005\u003c/li\u003e\n\u003cli\u003eDelgado-Morales, R., Agis-Balboa, R. C., Esteller, M. \u0026amp; Berdasco, M. Epigenetic mechanisms during ageing and neurogenesis as novel therapeutic avenues in human brain disorders. \u003cem\u003eClin Epigenetics\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 67 (2017). https://doi.org:10.1186/s13148-017-0365-z\u003c/li\u003e\n\u003cli\u003eNikolac Perkovic, M.\u003cem\u003e et al.\u003c/em\u003e Epigenetics of Alzheimer\u0026apos;s Disease. \u003cem\u003eBiomolecules\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e (2021). https://doi.org:10.3390/biom11020195\u003c/li\u003e\n\u003cli\u003eSen, P., Shah, P. P., Nativio, R. \u0026amp; Berger, S. L. Epigenetic Mechanisms of Longevity and Aging. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e166\u003c/strong\u003e, 822-839 (2016). https://doi.org:10.1016/j.cell.2016.07.050\u003c/li\u003e\n\u003cli\u003eWang, K.\u003cem\u003e et al.\u003c/em\u003e Epigenetic regulation of aging: implications for interventions of aging and diseases. \u003cem\u003eSignal Transduct Target Ther\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 374 (2022). https://doi.org:10.1038/s41392-022-01211-8\u003c/li\u003e\n\u003cli\u003eJeremic, D., Jimenez-Diaz, L. \u0026amp; Navarro-Lopez, J. D. Targeting epigenetics: A novel promise for Alzheimer\u0026apos;s disease treatment. \u003cem\u003eAgeing Res Rev\u003c/em\u003e \u003cstrong\u003e90\u003c/strong\u003e, 102003 (2023). https://doi.org:10.1016/j.arr.2023.102003\u003c/li\u003e\n\u003cli\u003eNativio, R.\u003cem\u003e et al.\u003c/em\u003e An integrated multi-omics approach identifies epigenetic alterations associated with Alzheimer\u0026apos;s disease. \u003cem\u003eNat Genet\u003c/em\u003e \u003cstrong\u003e52\u003c/strong\u003e, 1024-1035 (2020). https://doi.org:10.1038/s41588-020-0696-0\u003c/li\u003e\n\u003cli\u003eAlzheimer\u0026apos;s disease facts and figures. \u003cem\u003eAlzheimers Dement\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 3708-3821 (2024). https://doi.org:10.1002/alz.13809\u003c/li\u003e\n\u003cli\u003eJack, C. R., Jr.\u003cem\u003e et al.\u003c/em\u003e NIA-AA Research Framework: Toward a biological definition of Alzheimer\u0026apos;s disease. \u003cem\u003eAlzheimers Dement\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 535-562 (2018). https://doi.org:10.1016/j.jalz.2018.02.018\u003c/li\u003e\n\u003cli\u003eHeneka, M. T.\u003cem\u003e et al.\u003c/em\u003e Neuroinflammation in Alzheimer\u0026apos;s disease. \u003cem\u003eLancet Neurol\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 388-405 (2015). https://doi.org:10.1016/S1474-4422(15)70016-5\u003c/li\u003e\n\u003cli\u003ePalop, J. J. \u0026amp; Mucke, L. Amyloid-beta-induced neuronal dysfunction in Alzheimer\u0026apos;s disease: from synapses toward neural networks. \u003cem\u003eNat Neurosci\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 812-818 (2010). https://doi.org:10.1038/nn.2583\u003c/li\u003e\n\u003cli\u003eKoronyo-Hamaoui, M.\u003cem\u003e et al.\u003c/em\u003e Identification of amyloid plaques in retinas from Alzheimer\u0026apos;s patients and noninvasive in vivo optical imaging of retinal plaques in a mouse model. \u003cem\u003eNeuroimage\u003c/em\u003e \u003cstrong\u003e54 Suppl 1\u003c/strong\u003e, S204-217 (2011). https://doi.org:10.1016/j.neuroimage.2010.06.020\u003c/li\u003e\n\u003cli\u003eKoronyo, Y.\u003cem\u003e et al.\u003c/em\u003e Retinal amyloid pathology and proof-of-concept imaging trial in Alzheimer\u0026apos;s disease. \u003cem\u003eJCI Insight\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e (2017). https://doi.org:10.1172/jci.insight.93621\u003c/li\u003e\n\u003cli\u003eLa Morgia, C.\u003cem\u003e et al.\u003c/em\u003e Melanopsin retinal ganglion cell loss in Alzheimer disease. \u003cem\u003eAnn Neurol\u003c/em\u003e \u003cstrong\u003e79\u003c/strong\u003e, 90-109 (2016). https://doi.org:10.1002/ana.24548\u003c/li\u003e\n\u003cli\u003eKoronyo, Y.\u003cem\u003e et al.\u003c/em\u003e Retinal pathological features and proteome signatures of Alzheimer\u0026apos;s disease. \u003cem\u003eActa Neuropathol\u003c/em\u003e \u003cstrong\u003e145\u003c/strong\u003e, 409-438 (2023). https://doi.org:10.1007/s00401-023-02548-2\u003c/li\u003e\n\u003cli\u003eShi, H.\u003cem\u003e et al.\u003c/em\u003e Identification of early pericyte loss and vascular amyloidosis in Alzheimer\u0026apos;s disease retina. \u003cem\u003eActa Neuropathol\u003c/em\u003e \u003cstrong\u003e139\u003c/strong\u003e, 813-836 (2020). https://doi.org:10.1007/s00401-020-02134-w\u003c/li\u003e\n\u003cli\u003eShi, H.\u003cem\u003e et al.\u003c/em\u003e Retinal arterial Abeta(40) deposition is linked with tight junction loss and cerebral amyloid angiopathy in MCI and AD patients. \u003cem\u003eAlzheimers Dement\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 5185-5197 (2023). https://doi.org:10.1002/alz.13086\u003c/li\u003e\n\u003cli\u003eShi, H.\u003cem\u003e et al.\u003c/em\u003e Identification of retinal oligomeric, citrullinated, and other tau isoforms in early and advanced AD and relations to disease status. \u003cem\u003eActa Neuropathol\u003c/em\u003e \u003cstrong\u003e148\u003c/strong\u003e, 3 (2024). https://doi.org:10.1007/s00401-024-02760-8\u003c/li\u003e\n\u003cli\u003eGaire, B. P.\u003cem\u003e et al.\u003c/em\u003e Alzheimer\u0026apos;s disease pathophysiology in the Retina. \u003cem\u003eProg Retin Eye Res\u003c/em\u003e \u003cstrong\u003e101\u003c/strong\u003e, 101273 (2024). https://doi.org:10.1016/j.preteyeres.2024.101273\u003c/li\u003e\n\u003cli\u003eGrimaldi, A.\u003cem\u003e et al.\u003c/em\u003e Neuroinflammatory Processes, A1 Astrocyte Activation and Protein Aggregation in the Retina of Alzheimer\u0026apos;s Disease Patients, Possible Biomarkers for Early Diagnosis. \u003cem\u003eFront Neurosci\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 925 (2019). https://doi.org:10.3389/fnins.2019.00925\u003c/li\u003e\n\u003cli\u003eXu, Q. A.\u003cem\u003e et al.\u003c/em\u003e Muller cell degeneration and microglial dysfunction in the Alzheimer\u0026apos;s retina. \u003cem\u003eActa Neuropathol Commun\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 145 (2022). https://doi.org:10.1186/s40478-022-01448-y\u003c/li\u003e\n\u003cli\u003eWijesinghe, P.\u003cem\u003e et al.\u003c/em\u003e Decoding amyloid beta clearance systems at inner blood-retina barrier using three-dimensional ex vivo retinal imaging in Alzheimer\u0026apos;s disease. \u003cem\u003eAlzheimers Dement\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, e70592 (2025). https://doi.org:10.1002/alz.70592\u003c/li\u003e\n\u003cli\u003eWalkiewicz, G.\u003cem\u003e et al.\u003c/em\u003e Primary retinal tauopathy: A tauopathy with a distinct molecular pattern. \u003cem\u003eAlzheimers Dement\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 330-340 (2024). https://doi.org:10.1002/alz.13424\u003c/li\u003e\n\u003cli\u003eHart de Ruyter, F. J.\u003cem\u003e et al.\u003c/em\u003e Phosphorylated tau in the retina correlates with tau pathology in the brain in Alzheimer\u0026apos;s disease and primary tauopathies. \u003cem\u003eActa Neuropathol\u003c/em\u003e \u003cstrong\u003e145\u003c/strong\u003e, 197-218 (2023). https://doi.org:10.1007/s00401-022-02525-1\u003c/li\u003e\n\u003cli\u003eHinton, D. R., Sadun, A. A., Blanks, J. C. \u0026amp; Miller, C. A. Optic-nerve degeneration in Alzheimer\u0026apos;s disease. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e315\u003c/strong\u003e, 485-487 (1986). https://doi.org:10.1056/NEJM198608213150804\u003c/li\u003e\n\u003cli\u003eGaire, B. P.\u003cem\u003e et al.\u003c/em\u003e Identification of Chlamydia pneumoniae and NLRP3 inflammasome activation in Alzheimer\u0026apos;s disease retina. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 771 (2026). https://doi.org:10.1038/s41467-026-68580-4\u003c/li\u003e\n\u003cli\u003ePennington, K. L. \u0026amp; DeAngelis, M. M. Epigenetic Mechanisms of the Aging Human Retina. \u003cem\u003eJ Exp Neurosci\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 51-79 (2015). https://doi.org:10.4137/JEN.S25513\u003c/li\u003e\n\u003cli\u003eXu, C., Fu, X., Qin, H. \u0026amp; Yao, K. Traversing the epigenetic landscape: DNA methylation from retina to brain in development and disease. \u003cem\u003eFront Cell Neurosci\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 1499719 (2024). https://doi.org:10.3389/fncel.2024.1499719\u003c/li\u003e\n\u003cli\u003eAdvani, J.\u003cem\u003e et al.\u003c/em\u003e QTL mapping of human retina DNA methylation identifies 87 gene-epigenome interactions in age-related macular degeneration. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1972 (2024). https://doi.org:10.1038/s41467-024-46063-8\u003c/li\u003e\n\u003cli\u003eMondal, A. K., Gaur, M., Advani, J. \u0026amp; Swaroop, A. Epigenome-metabolism nexus in the retina: implications for aging and disease. \u003cem\u003eTrends Genet\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 718-729 (2024). https://doi.org:10.1016/j.tig.2024.04.012\u003c/li\u003e\n\u003cli\u003eBowman, G. D. \u0026amp; Poirier, M. G. Post-translational modifications of histones that influence nucleosome dynamics. \u003cem\u003eChem Rev\u003c/em\u003e \u003cstrong\u003e115\u003c/strong\u003e, 2274-2295 (2015). https://doi.org:10.1021/cr500350x\u003c/li\u003e\n\u003cli\u003eKouzarides, T. Chromatin modifications and their function. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e128\u003c/strong\u003e, 693-705 (2007). https://doi.org:10.1016/j.cell.2007.02.005\u003c/li\u003e\n\u003cli\u003eSultan, F. A. \u0026amp; Day, J. J. Epigenetic mechanisms in memory and synaptic function. \u003cem\u003eEpigenomics\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 157-181 (2011). https://doi.org:10.2217/epi.11.6\u003c/li\u003e\n\u003cli\u003eHerre, M. \u0026amp; Korb, E. The chromatin landscape of neuronal plasticity. \u003cem\u003eCurr Opin Neurobiol\u003c/em\u003e \u003cstrong\u003e59\u003c/strong\u003e, 79-86 (2019). https://doi.org:10.1016/j.conb.2019.04.006\u003c/li\u003e\n\u003cli\u003eSingh, P., Srivas, S. \u0026amp; Thakur, M. K. Epigenetic Regulation of Memory-Therapeutic Potential for Disorders. \u003cem\u003eCurr Neuropharmacol\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1208-1221 (2017). https://doi.org:10.2174/1570159X15666170404144522\u003c/li\u003e\n\u003cli\u003eCampbell, R. R. \u0026amp; Wood, M. A. How the epigenome integrates information and reshapes the synapse. \u003cem\u003eNat Rev Neurosci\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 133-147 (2019). https://doi.org:10.1038/s41583-019-0121-9\u003c/li\u003e\n\u003cli\u003eWalker, M. P., LaFerla, F. M., Oddo, S. S. \u0026amp; Brewer, G. J. Reversible epigenetic histone modifications and Bdnf expression in neurons with aging and from a mouse model of Alzheimer\u0026apos;s disease. \u003cem\u003eAge (Dordr)\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, 519-531 (2013). https://doi.org:10.1007/s11357-011-9375-5\u003c/li\u003e\n\u003cli\u003eLee, M. Y.\u003cem\u003e et al.\u003c/em\u003e Epigenome signatures landscaped by histone H3K9me3 are associated with the synaptic dysfunction in Alzheimer\u0026apos;s disease. \u003cem\u003eAging Cell\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, e13153 (2020). https://doi.org:10.1111/acel.13153\u003c/li\u003e\n\u003cli\u003eZheng, Y.\u003cem\u003e et al.\u003c/em\u003e Inhibition of EHMT1/2 rescues synaptic and cognitive functions for Alzheimer\u0026apos;s disease. \u003cem\u003eBrain\u003c/em\u003e \u003cstrong\u003e142\u003c/strong\u003e, 787-807 (2019). https://doi.org:10.1093/brain/awy354\u003c/li\u003e\n\u003cli\u003eCoppede, F. The potential of epigenetic therapies in neurodegenerative diseases. \u003cem\u003eFront Genet\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 220 (2014). https://doi.org:10.3389/fgene.2014.00220\u003c/li\u003e\n\u003cli\u003eAdwan, L. \u0026amp; Zawia, N. H. Epigenetics: a novel therapeutic approach for the treatment of Alzheimer\u0026apos;s disease. \u003cem\u003ePharmacol Ther\u003c/em\u003e \u003cstrong\u003e139\u003c/strong\u003e, 41-50 (2013). https://doi.org:10.1016/j.pharmthera.2013.03.010\u003c/li\u003e\n\u003cli\u003eBaumann, M.\u003cem\u003e et al.\u003c/em\u003e Tricyclic Analogues of Epidithiodioxopiperazine Alkaloids with Promising In Vitro and In Vivo Antitumor Activity. \u003cem\u003eChem Sci\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 4451-4457 (2015). https://doi.org:10.1039/C5SC01536G\u003c/li\u003e\n\u003cli\u003eFyodorov, D. V., Zhou, B. R., Skoultchi, A. I. \u0026amp; Bai, Y. Emerging roles of linker histones in regulating chromatin structure and function. \u003cem\u003eNat Rev Mol Cell Biol\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 192-206 (2018). https://doi.org:10.1038/nrm.2017.94\u003c/li\u003e\n\u003cli\u003eTvardovskiy, A., Schwammle, V., Kempf, S. J., Rogowska-Wrzesinska, A. \u0026amp; Jensen, O. N. Accumulation of histone variant H3.3 with age is associated with profound changes in the histone methylation landscape. \u003cem\u003eNucleic Acids Res\u003c/em\u003e \u003cstrong\u003e45\u003c/strong\u003e, 9272-9289 (2017). https://doi.org:10.1093/nar/gkx696\u003c/li\u003e\n\u003cli\u003eMaze, I.\u003cem\u003e et al.\u003c/em\u003e Critical Role of Histone Turnover in Neuronal Transcription and Plasticity. \u003cem\u003eNeuron\u003c/em\u003e \u003cstrong\u003e87\u003c/strong\u003e, 77-94 (2015). https://doi.org:10.1016/j.neuron.2015.06.014\u003c/li\u003e\n\u003cli\u003eGallegos, D. A., Chan, U., Chen, L. F. \u0026amp; West, A. E. Chromatin Regulation of Neuronal Maturation and Plasticity. \u003cem\u003eTrends Neurosci\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 311-324 (2018). https://doi.org:10.1016/j.tins.2018.02.009\u003c/li\u003e\n\u003cli\u003eOhzeki, J.\u003cem\u003e et al.\u003c/em\u003e KAT7/HBO1/MYST2 Regulates CENP-A Chromatin Assembly by Antagonizing Suv39h1-Mediated Centromere Inactivation. \u003cem\u003eDev Cell\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 413-427 (2016). https://doi.org:10.1016/j.devcel.2016.05.006\u003c/li\u003e\n\u003cli\u003eChang, B., Chen, Y., Zhao, Y. \u0026amp; Bruick, R. K. JMJD6 is a histone arginine demethylase. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e318\u003c/strong\u003e, 444-447 (2007). https://doi.org:10.1126/science.1145801\u003c/li\u003e\n\u003cli\u003eChapman, P. F.\u003cem\u003e et al.\u003c/em\u003e Impaired synaptic plasticity and learning in aged amyloid precursor protein transgenic mice. \u003cem\u003eNat Neurosci\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 271-276 (1999). https://doi.org:10.1038/6374\u003c/li\u003e\n\u003cli\u003eButterfield, D. A. Phosphoproteomics of Alzheimer disease brain: Insights into altered brain protein regulation of critical neuronal functions and their contributions to subsequent cognitive loss. \u003cem\u003eBiochim Biophys Acta Mol Basis Dis\u003c/em\u003e \u003cstrong\u003e1865\u003c/strong\u003e, 2031-2039 (2019). https://doi.org:10.1016/j.bbadis.2018.08.035\u003c/li\u003e\n\u003cli\u003eCardozo, P. L.\u003cem\u003e et al.\u003c/em\u003e Synaptic Elimination in Neurological Disorders. \u003cem\u003eCurr Neuropharmacol\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 1071-1095 (2019). https://doi.org:10.2174/1570159X17666190603170511\u003c/li\u003e\n\u003cli\u003eOddo, S.\u003cem\u003e et al.\u003c/em\u003e Triple-transgenic model of Alzheimer\u0026apos;s disease with plaques and tangles: intracellular Abeta and synaptic dysfunction. \u003cem\u003eNeuron\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, 409-421 (2003). https://doi.org:10.1016/s0896-6273(03)00434-3\u003c/li\u003e\n\u003cli\u003eButovsky, O.\u003cem\u003e et al.\u003c/em\u003e Glatiramer acetate fights against Alzheimer\u0026apos;s disease by inducing dendritic-like microglia expressing insulin-like growth factor 1. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e103\u003c/strong\u003e, 11784-11789 (2006). https://doi.org:10.1073/pnas.0604681103\u003c/li\u003e\n\u003cli\u003eDoustar, J.\u003cem\u003e et al.\u003c/em\u003e Parallels between retinal and brain pathology and response to immunotherapy in old, late-stage Alzheimer\u0026apos;s disease mouse models. \u003cem\u003eAging Cell\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, e13246 (2020). https://doi.org:10.1111/acel.13246\u003c/li\u003e\n\u003cli\u003eKasindi, A.\u003cem\u003e et al.\u003c/em\u003e Glatiramer Acetate Immunomodulation: Evidence of Neuroprotection and Cognitive Preservation. \u003cem\u003eCells\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e (2022). https://doi.org:10.3390/cells11091578\u003c/li\u003e\n\u003cli\u003eKoronyo, Y.\u003cem\u003e et al.\u003c/em\u003e Therapeutic effects of glatiramer acetate and grafted CD115(+) monocytes in a mouse model of Alzheimer\u0026apos;s disease. \u003cem\u003eBrain\u003c/em\u003e \u003cstrong\u003e138\u003c/strong\u003e, 2399-2422 (2015). https://doi.org:10.1093/brain/awv150\u003c/li\u003e\n\u003cli\u003eKoronyo-Hamaoui, M.\u003cem\u003e et al.\u003c/em\u003e Attenuation of AD-like neuropathology by harnessing peripheral immune cells: local elevation of IL-10 and MMP-9. \u003cem\u003eJ Neurochem\u003c/em\u003e \u003cstrong\u003e111\u003c/strong\u003e, 1409-1424 (2009). https://doi.org:10.1111/j.1471-4159.2009.06402.x\u003c/li\u003e\n\u003cli\u003eBernstein, K. E.\u003cem\u003e et al.\u003c/em\u003e Angiotensin-converting enzyme overexpression in myelomonocytes prevents Alzheimer\u0026apos;s-like cognitive decline. \u003cem\u003eJ Clin Invest\u003c/em\u003e \u003cstrong\u003e124\u003c/strong\u003e, 1000-1012 (2014). https://doi.org:10.1172/JCI66541\u003c/li\u003e\n\u003cli\u003eKoronyo-Hamaoui, M.\u003cem\u003e et al.\u003c/em\u003e Peripherally derived angiotensin converting enzyme-enhanced macrophages alleviate Alzheimer-related disease. \u003cem\u003eBrain\u003c/em\u003e \u003cstrong\u003e143\u003c/strong\u003e, 336-358 (2020). https://doi.org:10.1093/brain/awz364\u003c/li\u003e\n\u003cli\u003eLi, S.\u003cem\u003e et al.\u003c/em\u003e Activated Bone Marrow-Derived Macrophages Eradicate Alzheimer\u0026apos;s-Related Abeta(42) Oligomers and Protect Synapses. \u003cem\u003eFront Immunol\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 49 (2020). https://doi.org:10.3389/fimmu.2020.00049\u003c/li\u003e\n\u003cli\u003eAlves, V. C., Figueiro-Silva, J., Ferrer, I. \u0026amp; Carro, E. Epigenetic silencing of OR and TAS2R genes expression in human orbitofrontal cortex at early stages of sporadic Alzheimer\u0026apos;s disease. \u003cem\u003eCell Mol Life Sci\u003c/em\u003e \u003cstrong\u003e80\u003c/strong\u003e, 196 (2023). https://doi.org:10.1007/s00018-023-04845-1\u003c/li\u003e\n\u003cli\u003eGil, L.\u003cem\u003e et al.\u003c/em\u003e Pathological Nuclear Hallmarks in Dentate Granule Cells of Alzheimer\u0026apos;s Patients: A Biphasic Regulation of Neurogenesis. \u003cem\u003eInt J Mol Sci\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e (2022). https://doi.org:10.3390/ijms232112873\u003c/li\u003e\n\u003cli\u003eDavis, M. R.\u003cem\u003e et al.\u003c/em\u003e Retinal ganglion cell vulnerability to pathogenic tau in Alzheimer\u0026apos;s disease. \u003cem\u003eActa Neuropathol Commun\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 31 (2025). https://doi.org:10.1186/s40478-025-01935-y\u003c/li\u003e\n\u003cli\u003eSchultz, N., Byman, E., Netherlands Brain, B. \u0026amp; Wennstrom, M. Levels of Retinal Amyloid-beta Correlate with Levels of Retinal IAPP and Hippocampal Amyloid-beta in Neuropathologically Evaluated Individuals. \u003cem\u003eJ Alzheimers Dis\u003c/em\u003e \u003cstrong\u003e73\u003c/strong\u003e, 1201-1209 (2020). https://doi.org:10.3233/JAD-190868\u003c/li\u003e\n\u003cli\u003eSantiago, J.\u003cem\u003e et al.\u003c/em\u003e Retinal tau phosphorylation in Alzheimer\u0026apos;s disease: A mass spectrometry study. \u003cem\u003eNeurobiol Dis\u003c/em\u003e \u003cstrong\u003e215\u003c/strong\u003e, 107057 (2025). https://doi.org:10.1016/j.nbd.2025.107057\u003c/li\u003e\n\u003cli\u003eDumitrascu, O. M.\u003cem\u003e et al.\u003c/em\u003e Retinal peri-arteriolar versus peri-venular amyloidosis, hippocampal atrophy, and cognitive impairment: exploratory trial. \u003cem\u003eActa Neuropathol Commun\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 109 (2024). https://doi.org:10.1186/s40478-024-01810-2\u003c/li\u003e\n\u003cli\u003eDumitrascu, O. M.\u003cem\u003e et al.\u003c/em\u003e Sectoral segmentation of retinal amyloid imaging in subjects with cognitive decline. \u003cem\u003eAlzheimers Dement (Amst)\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, e12109 (2020). https://doi.org:10.1002/dad2.12109\u003c/li\u003e\n\u003cli\u003eSnigdha, S.\u003cem\u003e et al.\u003c/em\u003e H3K9me3 Inhibition Improves Memory, Promotes Spine Formation, and Increases BDNF Levels in the Aged Hippocampus. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 3611-3622 (2016). https://doi.org:10.1523/JNEUROSCI.2693-15.2016\u003c/li\u003e\n\u003cli\u003eBloss, E. B.\u003cem\u003e et al.\u003c/em\u003e Evidence for reduced experience-dependent dendritic spine plasticity in the aging prefrontal cortex. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 7831-7839 (2011). https://doi.org:10.1523/JNEUROSCI.0839-11.2011\u003c/li\u003e\n\u003cli\u003eMahmmoud, R. R.\u003cem\u003e et al.\u003c/em\u003e Spatial and Working Memory Is Linked to Spine Density and Mushroom Spines. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, e0139739 (2015). https://doi.org:10.1371/journal.pone.0139739\u003c/li\u003e\n\u003cli\u003eXu, B.\u003cem\u003e et al.\u003c/em\u003e Loss of thin spines and small synapses contributes to defective hippocampal function in aged mice. \u003cem\u003eNeurobiol Aging\u003c/em\u003e \u003cstrong\u003e71\u003c/strong\u003e, 91-104 (2018). https://doi.org:10.1016/j.neurobiolaging.2018.07.010\u003c/li\u003e\n\u003cli\u003eKarpova, N. N. Role of BDNF epigenetics in activity-dependent neuronal plasticity. \u003cem\u003eNeuropharmacology\u003c/em\u003e \u003cstrong\u003e76 Pt C\u003c/strong\u003e, 709-718 (2014). https://doi.org:10.1016/j.neuropharm.2013.04.002\u003c/li\u003e\n\u003cli\u003eGupta-Agarwal, S.\u003cem\u003e et al.\u003c/em\u003e G9a/GLP histone lysine dimethyltransferase complex activity in the hippocampus and the entorhinal cortex is required for gene activation and silencing during memory consolidation. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 5440-5453 (2012). https://doi.org:10.1523/JNEUROSCI.0147-12.2012\u003c/li\u003e\n\u003cli\u003eIonescu-Tucker, A.\u003cem\u003e et al.\u003c/em\u003e Exercise Reduces H3K9me3 and Regulates Brain Derived Neurotrophic Factor and GABRA2 in an Age Dependent Manner. \u003cem\u003eFront Aging Neurosci\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 798297 (2021). https://doi.org:10.3389/fnagi.2021.798297\u003c/li\u003e\n\u003cli\u003eAmidfar, M., de Oliveira, J., Kucharska, E., Budni, J. \u0026amp; Kim, Y. K. The role of CREB and BDNF in neurobiology and treatment of Alzheimer\u0026apos;s disease. \u003cem\u003eLife Sci\u003c/em\u003e \u003cstrong\u003e257\u003c/strong\u003e, 118020 (2020). https://doi.org:10.1016/j.lfs.2020.118020\u003c/li\u003e\n\u003cli\u003eJiao, S. S.\u003cem\u003e et al.\u003c/em\u003e Brain-derived neurotrophic factor protects against tau-related neurodegeneration of Alzheimer\u0026apos;s disease. \u003cem\u003eTransl Psychiatry\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, e907 (2016). https://doi.org:10.1038/tp.2016.186\u003c/li\u003e\n\u003cli\u003eO\u0026apos;Bryant, S. E.\u003cem\u003e et al.\u003c/em\u003e Brain-derived neurotrophic factor levels in Alzheimer\u0026apos;s disease. \u003cem\u003eJ Alzheimers Dis\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 337-341 (2009). https://doi.org:10.3233/JAD-2009-1051\u003c/li\u003e\n\u003cli\u003eSong, J. H., Yu, J. T. \u0026amp; Tan, L. Brain-Derived Neurotrophic Factor in Alzheimer\u0026apos;s Disease: Risk, Mechanisms, and Therapy. \u003cem\u003eMol Neurobiol\u003c/em\u003e \u003cstrong\u003e52\u003c/strong\u003e, 1477-1493 (2015). https://doi.org:10.1007/s12035-014-8958-4\u003c/li\u003e\n\u003cli\u003eLin, W. J.\u003cem\u003e et al.\u003c/em\u003e VGF and Its C-Terminal Peptide TLQP-62 Regulate Memory Formation in Hippocampus via a BDNF-TrkB-Dependent Mechanism. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, 10343-10356 (2015). https://doi.org:10.1523/JNEUROSCI.0584-15.2015\u003c/li\u003e\n\u003cli\u003eRouillard, A. D.\u003cem\u003e et al.\u003c/em\u003e The harmonizome: a collection of processed datasets gathered to serve and mine knowledge about genes and proteins. \u003cem\u003eDatabase (Oxford)\u003c/em\u003e \u003cstrong\u003e2016\u003c/strong\u003e (2016). https://doi.org:10.1093/database/baw100\u003c/li\u003e\n\u003cli\u003eTrani, E.\u003cem\u003e et al.\u003c/em\u003e Isolation and characterization of VGF peptides in rat brain. Role of PC1/3 and PC2 in the maturation of VGF precursor. \u003cem\u003eJ Neurochem\u003c/em\u003e \u003cstrong\u003e81\u003c/strong\u003e, 565-574 (2002). https://doi.org:10.1046/j.1471-4159.2002.00842.x\u003c/li\u003e\n\u003cli\u003eAlder, J.\u003cem\u003e et al.\u003c/em\u003e Brain-derived neurotrophic factor-induced gene expression reveals novel actions of VGF in hippocampal synaptic plasticity. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 10800-10808 (2003). \u003c/li\u003e\n\u003cli\u003eLi, C.\u003cem\u003e et al.\u003c/em\u003e Neuropeptide VGF C-Terminal Peptide TLQP-62 Alleviates Lipopolysaccharide-Induced Memory Deficits and Anxiety-like and Depression-like Behaviors in Mice: The Role of BDNF/TrkB Signaling. \u003cem\u003eACS Chem Neurosci\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 2005-2018 (2017). https://doi.org:10.1021/acschemneuro.7b00154\u003c/li\u003e\n\u003cli\u003eLin, W. J.\u003cem\u003e et al.\u003c/em\u003e An increase in VGF expression through a rapid, transcription-independent, autofeedback mechanism improves cognitive function. \u003cem\u003eTransl Psychiatry\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 383 (2021). https://doi.org:10.1038/s41398-021-01489-2\u003c/li\u003e\n\u003cli\u003eBehnke, J.\u003cem\u003e et al.\u003c/em\u003e Neuropeptide VGF Promotes Maturation of Hippocampal Dendrites That Is Reduced by Single Nucleotide Polymorphisms. \u003cem\u003eInt J Mol Sci\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e (2017). https://doi.org:10.3390/ijms18030612\u003c/li\u003e\n\u003cli\u003eThakker-Varia, S.\u003cem\u003e et al.\u003c/em\u003e VGF (TLQP-62)-induced neurogenesis targets early phase neural progenitor cells in the adult hippocampus and requires glutamate and BDNF signaling. \u003cem\u003eStem Cell Res\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 762-777 (2014). https://doi.org:10.1016/j.scr.2014.03.005\u003c/li\u003e\n\u003cli\u003eTakei, N. \u0026amp; Nawa, H. mTOR signaling and its roles in normal and abnormal brain development. \u003cem\u003eFront Mol Neurosci\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 28 (2014). https://doi.org:10.3389/fnmol.2014.00028\u003c/li\u003e\n\u003cli\u003eBartolomucci, A.\u003cem\u003e et al.\u003c/em\u003e The extended granin family: structure, function, and biomedical implications. \u003cem\u003eEndocr Rev\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 755-797 (2011). https://doi.org:10.1210/er.2010-0027\u003c/li\u003e\n\u003cli\u003eTasaki, S., Gaiteri, C., Mostafavi, S., De Jager, P. L. \u0026amp; Bennett, D. A. The Molecular and Neuropathological Consequences of Genetic Risk for Alzheimer\u0026apos;s Dementia. \u003cem\u003eFront Neurosci\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 699 (2018). https://doi.org:10.3389/fnins.2018.00699\u003c/li\u003e\n\u003cli\u003eBai, B.\u003cem\u003e et al.\u003c/em\u003e Deep Multilayer Brain Proteomics Identifies Molecular Networks in Alzheimer\u0026apos;s Disease Progression. \u003cem\u003eNeuron\u003c/em\u003e \u003cstrong\u003e105\u003c/strong\u003e, 975-991 e977 (2020). https://doi.org:10.1016/j.neuron.2019.12.015\u003c/li\u003e\n\u003cli\u003eCarrette, O.\u003cem\u003e et al.\u003c/em\u003e A panel of cerebrospinal fluid potential biomarkers for the diagnosis of Alzheimer\u0026apos;s disease. \u003cem\u003eProteomics\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 1486-1494 (2003). https://doi.org:10.1002/pmic.200300470\u003c/li\u003e\n\u003cli\u003eKhoonsari, P. E.\u003cem\u003e et al.\u003c/em\u003e Improved Differential Diagnosis of Alzheimer\u0026apos;s Disease by Integrating ELISA and Mass Spectrometry-Based Cerebrospinal Fluid Biomarkers. \u003cem\u003eJ Alzheimers Dis\u003c/em\u003e \u003cstrong\u003e67\u003c/strong\u003e, 639-651 (2019). https://doi.org:10.3233/JAD-180855\u003c/li\u003e\n\u003cli\u003ePedrero-Prieto, C. M.\u003cem\u003e et al.\u003c/em\u003e A comprehensive systematic review of CSF proteins and peptides that define Alzheimer\u0026apos;s disease. \u003cem\u003eClin Proteomics\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 21 (2020). https://doi.org:10.1186/s12014-020-09276-9\u003c/li\u003e\n\u003cli\u003eHoltta, M.\u003cem\u003e et al.\u003c/em\u003e An integrated workflow for multiplex CSF proteomics and peptidomics-identification of candidate cerebrospinal fluid biomarkers of Alzheimer\u0026apos;s disease. \u003cem\u003eJ Proteome Res\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 654-663 (2015). https://doi.org:10.1021/pr501076j\u003c/li\u003e\n\u003cli\u003eHendrickson, R. C.\u003cem\u003e et al.\u003c/em\u003e High Resolution Discovery Proteomics Reveals Candidate Disease Progression Markers of Alzheimer\u0026apos;s Disease in Human Cerebrospinal Fluid. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, e0135365 (2015). https://doi.org:10.1371/journal.pone.0135365\u003c/li\u003e\n\u003cli\u003eLlano, D. A., Bundela, S., Mudar, R. A., Devanarayan, V. \u0026amp; Alzheimer\u0026apos;s Disease Neuroimaging, I. A multivariate predictive modeling approach reveals a novel CSF peptide signature for both Alzheimer\u0026apos;s Disease state classification and for predicting future disease progression. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, e0182098 (2017). https://doi.org:10.1371/journal.pone.0182098\u003c/li\u003e\n\u003cli\u003eDuits, F. H.\u003cem\u003e et al.\u003c/em\u003e Synaptic proteins in CSF as potential novel biomarkers for prognosis in prodromal Alzheimer\u0026apos;s disease. \u003cem\u003eAlzheimers Res Ther\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 5 (2018). https://doi.org:10.1186/s13195-017-0335-x\u003c/li\u003e\n\u003cli\u003eSathe, G.\u003cem\u003e et al.\u003c/em\u003e Quantitative Proteomic Profiling of Cerebrospinal Fluid to Identify Candidate Biomarkers for Alzheimer\u0026apos;s Disease. \u003cem\u003eProteomics Clin Appl\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, e1800105 (2019). https://doi.org:10.1002/prca.201800105\u003c/li\u003e\n\u003cli\u003eBeckmann, N. D.\u003cem\u003e et al.\u003c/em\u003e Multiscale causal networks identify VGF as a key regulator of Alzheimer\u0026apos;s disease. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 3942 (2020). https://doi.org:10.1038/s41467-020-17405-z\u003c/li\u003e\n\u003cli\u003eSpellman, D. S.\u003cem\u003e et al.\u003c/em\u003e Development and evaluation of a multiplexed mass spectrometry based assay for measuring candidate peptide biomarkers in Alzheimer\u0026apos;s Disease Neuroimaging Initiative (ADNI) CSF. \u003cem\u003eProteomics Clin Appl\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 715-731 (2015). https://doi.org:10.1002/prca.201400178\u003c/li\u003e\n\u003cli\u003eJahn, H.\u003cem\u003e et al.\u003c/em\u003e Peptide fingerprinting of Alzheimer\u0026apos;s disease in cerebrospinal fluid: identification and prospective evaluation of new synaptic biomarkers. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, e26540 (2011). https://doi.org:10.1371/journal.pone.0026540\u003c/li\u003e\n\u003cli\u003eWang, X.\u003cem\u003e et al.\u003c/em\u003e Deciphering cellular transcriptional alterations in Alzheimer\u0026apos;s disease brains. \u003cem\u003eMol Neurodegener\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 38 (2020). https://doi.org:10.1186/s13024-020-00392-6\u003c/li\u003e\n\u003cli\u003ePark, S. A.\u003cem\u003e et al.\u003c/em\u003e SWATH-MS analysis of cerebrospinal fluid to generate a robust battery of biomarkers for Alzheimer\u0026apos;s disease. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 7423 (2020). https://doi.org:10.1038/s41598-020-64461-y\u003c/li\u003e\n\u003cli\u003eFargali, S.\u003cem\u003e et al.\u003c/em\u003e The granin VGF promotes genesis of secretory vesicles, and regulates circulating catecholamine levels and blood pressure. \u003cem\u003eFASEB J\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 2120-2133 (2014). https://doi.org:10.1096/fj.13-239509\u003c/li\u003e\n\u003cli\u003eGondre-Lewis, M. C., Park, J. J. \u0026amp; Loh, Y. P. Cellular mechanisms for the biogenesis and transport of synaptic and dense-core vesicles. \u003cem\u003eInt Rev Cell Mol Biol\u003c/em\u003e \u003cstrong\u003e299\u003c/strong\u003e, 27-115 (2012). https://doi.org:10.1016/B978-0-12-394310-1.00002-3\u003c/li\u003e\n\u003cli\u003eHariri, A. R.\u003cem\u003e et al.\u003c/em\u003e Brain-derived neurotrophic factor val66met polymorphism affects human memory-related hippocampal activity and predicts memory performance. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 6690-6694 (2003). https://doi.org:10.1523/JNEUROSCI.23-17-06690.2003\u003c/li\u003e\n\u003cli\u003eEgan, M. F.\u003cem\u003e et al.\u003c/em\u003e The BDNF val66met polymorphism affects activity-dependent secretion of BDNF and human memory and hippocampal function. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e112\u003c/strong\u003e, 257-269 (2003). https://doi.org:10.1016/s0092-8674(03)00035-7\u003c/li\u003e\n\u003cli\u003eBui, T. M., Wiesolek, H. L. \u0026amp; Sumagin, R. ICAM-1: A master regulator of cellular responses in inflammation, injury resolution, and tumorigenesis. \u003cem\u003eJ Leukoc Biol\u003c/em\u003e \u003cstrong\u003e108\u003c/strong\u003e, 787-799 (2020). https://doi.org:10.1002/JLB.2MR0220-549R\u003c/li\u003e\n\u003cli\u003eMaier, M.\u003cem\u003e et al.\u003c/em\u003e Complement C3 deficiency leads to accelerated amyloid beta plaque deposition and neurodegeneration and modulation of the microglia/macrophage phenotype in amyloid precursor protein transgenic mice. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 6333-6341 (2008). https://doi.org:10.1523/JNEUROSCI.0829-08.2008\u003c/li\u003e\n\u003cli\u003eFrenkel, D., Maron, R., Burt, D. S. \u0026amp; Weiner, H. L. Nasal vaccination with a proteosome-based adjuvant and glatiramer acetate clears beta-amyloid in a mouse model of Alzheimer disease. \u003cem\u003eJ Clin Invest\u003c/em\u003e \u003cstrong\u003e115\u003c/strong\u003e, 2423-2433 (2005). https://doi.org:10.1172/JCI23241\u003c/li\u003e\n\u003cli\u003eBakalash, S.\u003cem\u003e et al.\u003c/em\u003e Egr1 expression is induced following glatiramer acetate immunotherapy in rodent models of glaucoma and Alzheimer\u0026apos;s disease. \u003cem\u003eInvest Ophthalmol Vis Sci\u003c/em\u003e \u003cstrong\u003e52\u003c/strong\u003e, 9033-9046 (2011). https://doi.org:10.1167/iovs.11-7498\u003c/li\u003e\n\u003cli\u003eZuroff, L., Daley, D., Black, K. L. \u0026amp; Koronyo-Hamaoui, M. Clearance of cerebral Abeta in Alzheimer\u0026apos;s disease: reassessing the role of microglia and monocytes. \u003cem\u003eCell Mol Life Sci\u003c/em\u003e \u003cstrong\u003e74\u003c/strong\u003e, 2167-2201 (2017). https://doi.org:10.1007/s00018-017-2463-7\u003c/li\u003e\n\u003cli\u003eLebson, L.\u003cem\u003e et al.\u003c/em\u003e Trafficking CD11b-positive blood cells deliver therapeutic genes to the brain of amyloid-depositing transgenic mice. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 9651-9658 (2010). https://doi.org:10.1523/JNEUROSCI.0329-10.2010\u003c/li\u003e\n\u003cli\u003eTheriault, P., ElAli, A. \u0026amp; Rivest, S. The dynamics of monocytes and microglia in Alzheimer\u0026apos;s disease. \u003cem\u003eAlzheimers Res Ther\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 41 (2015). https://doi.org:10.1186/s13195-015-0125-2\u003c/li\u003e\n\u003cli\u003eRosenzweig, N.\u003cem\u003e et al.\u003c/em\u003e PD-1/PD-L1 checkpoint blockade harnesses monocyte-derived macrophages to combat cognitive impairment in a tauopathy mouse model. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 465 (2019). https://doi.org:10.1038/s41467-019-08352-5\u003c/li\u003e\n\u003cli\u003eMunoz-Castro, C.\u003cem\u003e et al.\u003c/em\u003e Monocyte-derived cells invade brain parenchyma and amyloid plaques in human Alzheimer\u0026apos;s disease hippocampus. \u003cem\u003eActa Neuropathol Commun\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 31 (2023). https://doi.org:10.1186/s40478-023-01530-z\u003c/li\u003e\n\u003cli\u003eDeczkowska, A., Amit, I. \u0026amp; Schwartz, M. Microglial immune checkpoint mechanisms. \u003cem\u003eNat Neurosci\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 779-786 (2018). https://doi.org:10.1038/s41593-018-0145-x\u003c/li\u003e\n\u003cli\u003eButovsky, O., Kunis, G., Koronyo-Hamaoui, M. \u0026amp; Schwartz, M. Selective ablation of bone marrow-derived dendritic cells increases amyloid plaques in a mouse Alzheimer\u0026apos;s disease model. \u003cem\u003eEur J Neurosci\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 413-416 (2007). https://doi.org:10.1111/j.1460-9568.2007.05652.x\u003c/li\u003e\n\u003cli\u003eRentsendorj, A.\u003cem\u003e et al.\u003c/em\u003e A novel role for osteopontin in macrophage-mediated amyloid-beta clearance in Alzheimer\u0026apos;s models. \u003cem\u003eBrain Behav Immun\u003c/em\u003e \u003cstrong\u003e67\u003c/strong\u003e, 163-180 (2018). https://doi.org:10.1016/j.bbi.2017.08.019\u003c/li\u003e\n\u003cli\u003eHelwig, M.\u003cem\u003e et al.\u003c/em\u003e The neuroendocrine protein 7B2 suppresses the aggregation of neurodegenerative disease-related proteins. \u003cem\u003eJ Biol Chem\u003c/em\u003e \u003cstrong\u003e288\u003c/strong\u003e, 1114-1124 (2013). https://doi.org:10.1074/jbc.M112.417071\u003c/li\u003e\n\u003cli\u003eHoshino, A.\u003cem\u003e et al.\u003c/em\u003e A novel function for proSAAS as an amyloid anti-aggregant in Alzheimer\u0026apos;s disease. \u003cem\u003eJ Neurochem\u003c/em\u003e \u003cstrong\u003e128\u003c/strong\u003e, 419-430 (2014). https://doi.org:10.1111/jnc.12454\u003c/li\u003e\n\u003cli\u003eEl Gaamouch, F.\u003cem\u003e et al.\u003c/em\u003e VGF-derived peptide TLQP-21 modulates microglial function through C3aR1 signaling pathways and reduces neuropathology in 5xFAD mice. \u003cem\u003eMol Neurodegener\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 4 (2020). https://doi.org:10.1186/s13024-020-0357-x\u003c/li\u003e\n\u003cli\u003eElmadany, N.\u003cem\u003e et al.\u003c/em\u003e The VGF-derived Peptide TLQP21 Impairs Purinergic Control of Chemotaxis and Phagocytosis in Mouse Microglia. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 3320-3331 (2020). https://doi.org:10.1523/JNEUROSCI.1458-19.2020\u003c/li\u003e\n\u003cli\u003eCho, K.\u003cem\u003e et al.\u003c/em\u003e TLQP-21 mediated activation of microglial BV2 cells promotes clearance of extracellular fibril amyloid-beta. \u003cem\u003eBiochem Biophys Res Commun\u003c/em\u003e \u003cstrong\u003e524\u003c/strong\u003e, 764-771 (2020). https://doi.org:10.1016/j.bbrc.2020.01.111\u003c/li\u003e\n\u003cli\u003eLee, J., Han, M., Wang, K., Butler, L. R. \u0026amp; Sinclair, D. A. Epigenetic reprogramming for ocular aging and disease: Mechanisms, biomarkers, and the road to the clinic. \u003cem\u003eProg Retin Eye Res\u003c/em\u003e \u003cstrong\u003e111\u003c/strong\u003e, 101442 (2026). https://doi.org:10.1016/j.preteyeres.2026.101442\u003c/li\u003e\n\u003cli\u003ePelzel, H. R. \u0026amp; Nickells, R. W. A role for epigenetic changes in the development of retinal neurodegenerative conditions. \u003cem\u003eJ Ocul Biol Dis Infor\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 104-110 (2011). https://doi.org:10.1007/s12177-012-9079-9\u003c/li\u003e\n\u003cli\u003eVit, J. P.\u003cem\u003e et al.\u003c/em\u003e Color and contrast vision in mouse models of aging and Alzheimer\u0026apos;s disease using a novel visual-stimuli four-arm maze. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 1255 (2021). https://doi.org:10.1038/s41598-021-80988-0\u003c/li\u003e\n\u003cli\u003eVit, J. P.\u003cem\u003e et al.\u003c/em\u003e Visual-stimuli Four-arm Maze test to Assess Cognition and Vision in Mice. \u003cem\u003eBio Protoc\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, e4234 (2021). https://doi.org:10.21769/BioProtoc.4234\u003c/li\u003e\n\u003cli\u003eRisher, W. C., Ustunkaya, T., Singh Alvarado, J. \u0026amp; Eroglu, C. Rapid Golgi analysis method for efficient and unbiased classification of dendritic spines. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, e107591 (2014). https://doi.org:10.1371/journal.pone.0107591\u003c/li\u003e\n\u003cli\u003eMetsalu, T. \u0026amp; Vilo, J. ClustVis: a web tool for visualizing clustering of multivariate data using Principal Component Analysis and heatmap. \u003cem\u003eNucleic Acids Res\u003c/em\u003e \u003cstrong\u003e43\u003c/strong\u003e, W566-570 (2015). https://doi.org:10.1093/nar/gkv468\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1. Demographic and brain neuropathology data of human subjects used for cerebral and retinal histopathological analysis.\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 307px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDiagnosis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 169px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOne-Way ANOVA\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCN\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMCI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eF-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eP-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSubjects\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003cp\u003e4F/2M\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003cp\u003e4F/3M\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e13\u003c/p\u003e\n \u003cp\u003e7F/6M\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003eN/A\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003eN/A\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003e91.3 \u0026plusmn; 2.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e91.1 \u0026plusmn; 1.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e85.5 \u0026plusmn; 2.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e1.612\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.2212\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRace\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003e2H/4W\u003c/p\u003e\n \u003cp\u003e33/67%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e1B/6W\u003c/p\u003e\n \u003cp\u003e14/86%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e3A+B/10W\u003c/p\u003e\n \u003cp\u003e23/77%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003eN/A\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003eN/A\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMMSE\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003e28.2 \u0026plusmn; 1.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e22.8 \u0026plusmn; 2.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e16.3 \u0026plusmn; 1.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e9.601\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.0012\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCDR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003e0.17 \u0026plusmn; 0.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e1.50 \u0026plusmn; 0.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e2.27 \u0026plusmn; 0.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e10.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.0005\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBraak stage\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eStage I-II\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eStage III-IV\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eStage V-VI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003e2.7 \u0026plusmn; 0.7\u003c/p\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e3.6 \u0026plusmn; 0.6\u003c/p\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e5.2 \u0026plusmn; 0.2\u003c/p\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003cp\u003e11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e8.674\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.0016\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eABC score\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003e1.95 \u0026plusmn; 0.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e2.06 \u0026plusmn; 0.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e2.75 \u0026plusmn; 0.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e6.472\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.0059\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eA\u0026beta; score (A9)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003e1.95 \u0026plusmn; 0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e2.11 \u0026plusmn; 0.42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e3.82 \u0026plusmn; 0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e5.340\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.0133\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNFT score (A9)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003e0.08 \u0026plusmn; 0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e1.21 \u0026plusmn; 0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e1.82 \u0026plusmn; 0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e2.680\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.0919\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAtrophy score (A9)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003e1.50 \u0026plusmn; 0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 102px;\"\u003e\n \u003cp\u003e1.57 \u0026plusmn; 0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e2.50 \u0026plusmn; 0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e1.219\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.3148\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eWhere applicable, data are presented as mean \u0026plusmn; SEM. CN\u0026mdash;cognitively normal, MCI\u0026mdash;mild cognitive impairment, AD\u0026mdash;Alzheimer\u0026rsquo;s disease. F\u0026mdash;female, M\u0026mdash;male. Age at death in years. H\u0026mdash;Hispanic, W\u0026mdash;White, B\u0026mdash;Black, A\u0026mdash;Asian. MMSE\u0026mdash;mini mental state examination. CDR\u0026mdash;clinical dementia rating: 0\u0026mdash;normal cognition, 1\u0026mdash;mild dementia, 2\u0026mdash;moderate dementia, 3\u0026mdash;severe dementia. ABC score: A\u0026mdash;A\u0026beta; plaque score modified from Thal, B\u0026mdash;NFT stage modified from Braak, C\u0026mdash;Neuritic plaque score modified from CERAD. A\u0026beta;, NFT and atrophy severity scores: 0\u0026mdash;none, 1\u0026mdash;sparse, 3\u0026mdash;moderate, 5\u0026mdash;frequent. A9\u0026mdash;Brodmann area A9, dorsolateral prefrontal cortex. N/A\u0026mdash;not applicable.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Neurodegenerative disease, heterochromatin, histone modification, NT1721, ocular manifestations, granin family members","lastPublishedDoi":"10.21203/rs.3.rs-8913130/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8913130/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Epigenetic dysregulation is increasingly linked to ageing and neurodegeneration, yet its contribution to retinal and brain pathology in Alzheimer's disease (AD) remains uncharacterized. We quantified the repressive histone mark H3K9me3 in post-mortem retinas and brains from donors spanning normal cognition, mild cognitive impairment, and AD dementia. In both tissues, H3K9me3 increased in early stages and further in AD dementia, strongly associating with cognitive status and neuropathological burden. To assess causality, we inhibited the H3K9 methyltransferase SUV39H1 in APPswe/PS1dE9 and APPswe/tauP301L/PS1tm1Mpm mouse models. SUV39H1 inhibition lowered H3K9me3, mitigated AD-like pathology, restored synaptic integrity, and improved cognitive and visual performance. Proteomics revealed that H3K9me3 derepression reestablished retinal and brain proteostasis and promoted neuroprotection through immunomodulatory pathways and BDNF/VGF–granin signalling. These findings identify H3K9me3 as shared epigenetic driver of AD-related dysfunction, highlight H3K9me3 reduction as therapeutic strategy, and position the retina as an accessible extension of the brain for epigenetic studies.","manuscriptTitle":"H3K9me3 inhibition reverses Alzheimer′s progression by restoring synaptic and immune proteostasis across the brain–retina axis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-13 04:58:01","doi":"10.21203/rs.3.rs-8913130/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e9a77052-7003-42c8-b1ef-854f2912915f","owner":[],"postedDate":"March 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":64416372,"name":"Biological sciences/Neuroscience/Epigenetics in the nervous system/Epigenetics and behaviour"},{"id":64416373,"name":"Biological sciences/Neuroscience/Epigenetics in the nervous system/Epigenetics and plasticity"}],"tags":[],"updatedAt":"2026-03-13T04:58:03+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-13 04:58:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8913130","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8913130","identity":"rs-8913130","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}