Abstract
Introduction:
Visual dysfunction and retinal structural changes can occur early in Alzheimer's disease (AD), but the molecular basis of these early alterations remains unclear.
Methods:
We performed quantitative proteomic profiling of the retina and brain from 4-week-old triple-transgenic AD (3xTg-AD) mice carrying human PS1M146V, APPSwe, and tauP301L mutations, prior to detectable retinal morphological abnormalities.
Results:
Retinal morphology was normal in 4-week-old 3xTg-AD. Proteomic analysis identified 92 significantly altered proteins in the retina and 130 in the brain, with eight proteins overlapping between tissues. These overlapping proteins included three hemoglobin subunits and five proteins involved in protein homeostasis and vesicular transport. The retinal proteome was characterized by reduced vision-related proteins, altered small-molecule transporters, and decreased levels of proteins involved in mitochondrial energetics. In the brain, prominent changes were observed in mitochondrial proteins, including respiratory chain components and mitochondrial ribosomal subunits, as well as proteins linked to autophagy and synaptic vesicle pathways.
Discussion:
These findings identify early, common and tissue-specific proteomic changes in the retina and brain of 3xTg-AD mice prior to detectable retinal structure abnormalities. The data indicate early changes in proteins related to mitochondrial function and intracellular transport and support the use of retina as an accessible tissue for detecting preclinical AD pathology.
Introduction
Alzheimer’s disease (AD) is a progressive neurodegenerative disorder and the leading cause of dementia worldwide (; ). Currently, there is no effective treatment to halt or reverse disease progression, and interventions started after clinical onset may contribute to the high failure of clinical trials (; ; ). These limitations underscore the need to identify early biomarkers and define disease mechanisms during the preclinical stage of AD. The classical neuropathological hallmarks of AD are extracellular amyloid-β (Aβ) plaques and intracellular neurofibrillary tangles composed of hyperphosphorylated tau protein (). Aβ peptides, predominantly Aβ40 and the more aggregation-prone Aβ42, are generated from amyloid precursor protein (APP) through sequential proteolytic cleavage by β- and γ-secretases. Mutations in presenilin-1 (PS1), a core component of the γ-secretase complex, alter APP processing and preferentially increase Aβ42 production, thereby accelerating Aβ accumulation, synaptic dysfunction, and cognitive decline (; ; ; ). However, AD is now recognized as a multifactorial disorder that extends beyond Aβ and tau pathology. Emerging evidence highlights early mitochondrial and metabolic dysfunction, impaired proteostasis, neuroinflammation, and vascular contributions as integral components of disease progression (; ; ; ; ). Importantly, AD-associated pathology is not restricted to the brain but also affects other regions of the central nervous system (CNS), particularly the visual system (; ; ).
Visual dysfunction is increasingly recognized as an early feature of AD, preceding the onset of cognitive symptoms (; ). The retina, an accessible extension of the CNS, provides a unique window to assess the brain pathology (; ). Structural and functional retinal abnormalities have been reported in AD patients, including thinning of the retinal nerve fiber layer (RNFL) and ganglion cell layer (GCL), reduced contrast sensitivity, and altered pupillary responses (; ; ). Importantly, some of these retinal alterations correlate with cortical Aβ burden and cognitive impairment (; ), suggesting that retinal changes may mirror early neurodegenerative processes occurring in the brain. The retina is among the most metabolically active tissues in the body and is highly dependent on mitochondrial function and efficient nutrient transport to maintain visual processing (Wong-Riley, 2010; ; ). This high bioenergetic demand may render retinal tissue particularly sensitive to early metabolic and mitochondrial disturbances that characterize AD pathogenesis (Yaffe, 2004; ; Yang et al., 2022; Zhang et al., 2025).
To date, studies investigating retinal involvement in AD have largely emphasized structural and functional assessments, such as optical coherence tomography and electroretinography (; ; ; ). However, early molecular alterations in the retina during AD progression remain poorly characterized. Furthermore, it remains unclear if the retina shows similar early biochemical alterations as the brain. To this end, we investigated early proteomic changes in the retina and brain using triple-transgenic AD (3xTg-AD) mouse model, harboring PS1M146V, APPSwe, and tauP301L mutations (). In this model, detectable brain pathology typically emerges around 4–6 months of age, with intracellular Aβ accumulation beginning at 3–4 months and extracellular plaques appearing at 6–9 months, whereas 4-week-old mice do not yet show functional deficits (; ). We therefore focused on 4-week-old 3xTg-AD mice to capture early, pre-symptomatic alterations. We assessed retinal morphology and performed comprehensive quantitative proteomic analyses on both retina and brain tissues to identify early proteomic signatures in AD.
Methods
Animals
Female 3xTg-AD on a B6; 129 genetic background (JAX #034830) and age-matched wild-type (WT) controls on a B6; 129 background (JAX #101045) were all purchased from the Jackson Laboratory. Mice were housed and aged in the West Virginia University vivarium under a 12-h light/12-h dark cycle and maintained on a standard chow diet (2018 Teklad Global 18% Protein, Inotiv Inc.) and water available ad libitum. All animal procedures were conducted in compliance with the National Institutes of Health guidelines and ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Experimental protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of West Virginia University. Animals were genotyped by Transnetyx using polymerase chain reaction (PCR) analysis of genomic DNA isolated from tail samples.
Hematoxylin and eosin staining and analysis
Mice were sacrificed at 4 weeks and 28 weeks of age, and their eyes were carefully enucleated and immediately fixed in 1 mL of Excalibur’s Alcoholic Z-Fix (Excalibur Pathology Inc., Norman, OK). Tissues were subsequently processed, paraffin-embedded, sectioned, and stained with hematoxylin and eosin (H&E) by Excalibur Pathology. H&E-stained retinal sections were imaged using a Nikon Eclipse Ti microscope equipped with a DS-Ri2 camera (Nikon Instruments, Melville, NY). Thickness of the outer segments (OS), inner segments (IS), outer nuclear layer (ONL), and inner nuclear layer (INL) was quantified using ImageJ at six predefined locations across the retina near the optic nerve head (ONH) on both the inferior and superior sides of the eye, as previously described (). Briefly, three positions were defined on each side of the ONH (−1, −2, −3 and +1, +2, +3), with −1 and +1 located 20 nuclei away from the ONH, −3 and +3 located 20 nuclei inward from the peripheral retina, and −2 and +2 defined as intermediate points between these locations. Each measurement was replicated ten nuclei away from the first measurement. For each eye, ten sections were analyzed and values were averaged to obtain the final thickness.
Tissue collection and protein extraction
Mice at 4 weeks of age were euthanized, and the eyecups were immediately enucleated (N = 4 eyecups per group). Retinal dissections were performed in cold Hanks' Balanced Salt Solution (HBSS) to isolate the neural retina, as previously described (Zhu et al., 2018; Xu et al., 2020). Whole brains were excised and briefly rinsed in cold Phosphate Buffered Saline (PBS). All tissues were flash-frozen in liquid nitrogen and stored at −80 °C until further processing. For protein extraction, retina and brain samples were homogenized in radioimmunoprecipitation assay (RIPA) buffer supplemented with protease and phosphatase inhibitors (5 mg/mL). Homogenates were centrifuged at 12,000 rpm for 15 min at 4 °C, and the supernatants containing soluble proteins were collected. Protein concentrations were determined using the Pierce™ BCA Protein Assay Kit (see Table 1), and equal amounts of total protein were used for downstream digestion and mass spectrometry analysis.
TABLE 1
| Mouse strain | Catalog | Company | Location |
|---|---|---|---|
| 3xTg-AD | MMRRC_034830-JAX | The Jackson Laboratory | Bar Harbor, ME, United States of America |
| B6129SF1/J | IMSR_JAX:101043 | The Jackson Laboratory | Bar Harbor, ME, United States of America |
| Mouse diet | Catalog | Company | Location |
| 2018 Teklad Global 18% Protein Rodent Diet | 2018 | Inotiv Inc | West Lafayette, IN, United States of America |
| Reagents | Catalog | Company | Location |
| RIPA Lysis and Extraction Buffer | 89901 | Thermo Fisher Scientific | Rockford, IL United States of America |
| Pierce Protease and Phosphatase Inhibitor Mini Tablets | A32959 | Thermo Fisher Scientific | Rockford, IL United States of America |
| Pierce™ BCA Protein Assay Kit | 23225 | Thermo Fisher Scientific | Rockford, IL United States of America |
| Excalibur’s Alcoholic Z-Fix | 1024 | Excalibur Pathology Inc | Norman, OK United States of America |
| Phosphate Buffer Saline | 3013794 | Thermo Fisher Scientific | St. Louis, MO United States of America |
| Hanks’ Balanced Salt Solution | 2276814 | Thermo Fisher Scientific | Waltham, MA United States of America |
Key resources.
Quantitative proteomics
Total proteins from each sample were reduced, alkylated, and purified by chloroform/methanol extraction prior to digestion with sequencing grade modified porcine trypsin (Promega). Tryptic peptides were then separated by reverse phase XSelect CSH C18 2.5 um resin (Waters) on an in-line 150 × 0.075 mm column using an UltiMate 3,000 RSLCnano system (Thermo). Peptides were eluted using a 60 min gradient from 98:2 to 65:35 buffer A:B ratio (Buffer A contains 0.1% formic acid, 0.5% acetonitrile. Buffer B contains 0.1% formic acid, 99.9% acetonitrile). Eluted peptides were ionized by electrospray (2.2 kV) followed by mass spectrometric analysis on an Orbitrap Exploris 480 mass spectrometer (Thermo). To assemble a chromatogram library, six gas-phase fractions were acquired on the Orbitrap Exploris with 4 m/z DIA spectra (4 m/z precursor isolation windows at 30,000 resolution, normalized AGC target 100%, maximum inject time 66 m) using a staggered window pattern from narrow mass ranges using optimized window placements. Precursor spectra were acquired after each DIA duty cycle, spanning the m/z range of the gas-phase fraction (i.e. 496-602 m/z, 60,000 resolution, normalized AGC target 100%, maximum injection time 50 m). For wide-window acquisitions, the Orbitrap Exploris was configured to acquire a precursor scan (385–1015 m/z, 60,000 resolution, normalized AGC target 100%, maximum injection time 50 m) followed by 50 × 12 m/z DIA spectra (12 m/z precursor isolation windows at 15,000 resolution, normalized AGC target 100%, maximum injection time 33 m) using a staggered window pattern with optimized window placements. Precursor spectra were acquired after each DIA duty cycle.
Proteomics data processing
Following data acquisition, data were searched using an empirically corrected library and a quantitative analysis was performed to obtain a comprehensive proteomic profile. Proteins were identified and quantified using EncyclopeDIA () and visualized with Scaffold DIA using 1% false discovery thresholds at both the protein and peptide level. Protein intensity values were normalized across samples using an in-house ProteiNorm pipeline (), a tool for systematic evaluation of normalization methods, imputation of missing values, and comparisons of multiple differential abundance methods. Normalization methods evaluated included log2 normalization (Log2), median normalization (Median), mean normalization (Mean), variance stabilizing normalization (VSN) (), quantile normalization (Quantile) (), cyclic loess normalization (Cyclic Loess) (), global robust linear regression normalization (RLR), and global intensity normalization (Global Intensity) (). The individual performance of each method was evaluated by comparing of the following metrices: total intensity, pooled intragroup coefficient of variation (PCV), pooled intragroup median absolute deviation (PMAD), pooled intragroup estimate of variance (PEV), intragroup correlation, sample correlation heatmap (Pearson), and log2-ratio distributions. The normalized data were used to perform statistical analysis using linear models for microarray data (limma) with empirical Bayes (eBayes) smoothing to the standard errors (). Proteins with an FDR adjusted p-value <0.05 and a fold change (FC) > 2 were considered significant. Data are available via PRoteomic IDEntifications Database (PRIDE) with identifier PXD069995.
Bioinformatics analysis
Proteomics quantification results were further processed using bioinformatic platforms. Principal Component Analysis (PCA) and Variable Importance in Projection (VIP) scoring were performed using MetaboAnalyst (https://www.metaboanalyst.ca). Functional enrichment analysis of biological processes and cellular components for significantly changed proteins between 3xTg-AD and WT tissues, were conducted using the DAVID Bioinformatics Resources (https://davidbioinformatics.nih.gov). Protein-protein interaction network analysis was carried out using Metascape (https://metascape.org). Heatmaps and quantitative data visualizations were generated using GraphPad Prism (v9.5.1, GraphPad Software Inc.). To identify AD-associated proteins enriched in the human brain proteomic dataset, we queried the Agora AD Knowledge Portal (https://agora.adknowledgeportal.org). Identified associations were further validated through manual curation of published literature using PubMed, with combinations of relevant keywords (e.g., “[protein name] AND [Alzheimer’s disease OR neurodegenerative disease]”).
Statistics
The statistical analyses of retinal thickness are presented as the mean ± standard deviation (SD). Unpaired two-tailed t tests were performed in GraphPad Prism v9.4.1 to determine significance. Values were considered significant.
Results
Age-dependent retinal thinning in 3xTg-AD mice
To examine retinal morphological changes, we quantified the thickness of individual retinal layers in H&E-stained sections at 4-week-old and 28-week-old mice. At 4-week, retinal structure in 3xTg-AD mice was comparable to that of WT controls, with no differences across retinal layers (Figures 1A, 2A,C–F). At 28 weeks, selective reductions in retinal layer thickness were observed in 3xTg-AD mice at specific locations, affecting OS, IS, ONL, and INL (Figures 2G–J). This progressive decline in retinal thickness with age in 3xTg-AD mice is consistent with reports of retinal layer thinning in AD patients (; ; ).
FIGURE 1
FIGURE 2
Early retinal proteomic alterations in 3xTg-AD mice
To investigate early proteomic alterations preceding AD pathology, we performed quantitative proteomic analysis of retinas from 3xTg-AD and WT mice at 4 weeks of age (Figure 1B). Principal component analysis (PCA) showed a clear separation between two groups, indicating early global proteomic change in the retina from 3xTg-AD mice (Figure 3A). A total of 92 proteins were significantly altered in 3xTg-AD retinas, including 51 downregulated and 41 upregulated proteins (Figure 3B; Supplementary Table S1). Variable Importance in Projection (VIP) analysis identified top 15 proteins driving PCA group separation (Figure 3C) with prominent changes of mitochondrial proteins involved in energy metabolism, such as succinate-CoA ligase subunit β (SUCB2) and ADP/ATP translocase 4 (ADT4), as well as hemoglobin-related proteins (HBB1, HBB2, and A8DUK4).
FIGURE 3
Specifically, proteins associated with retinal structure and neuronal connectivity, including ermin (ERMIN) and plexin-A4 (PLXA4), whereas multiple histone H2B variants (H2B1M, H2B1C, H2B1H, H2B2B) and the histone demethylase JMJD6 were increased (Figure 3C). Gene ontology (GO) analysis showed enrichment of pathways related to transport, in particular oxygen transport, and vision (Figures 3D,E; Supplementary Table S2). Several solute carrier (SLC) transporters were differentially expressed in 3xTg-AD retina. SC5A3 was significantly upregulated, whereas ADT4, S6A13, and S22A8 were downregulated (Figure 3E). Notably, ADT4, a mitochondrial ADP/ATP translocase, was reduced by more than 11-fold, suggesting impaired mitochondrial energy output. In addition, proteins involved in vesicular transport were broadly altered. Components associated with endoplasmic reticulum (ER)-Golgi trafficking and synaptic vesicle dynamics, including TBC23, SNP23, SCFD2, USE1, and SFT2C, were decreased. In contrast, expression of proteins involved in Golgi organization, calcium-dependent synaptic signaling, and vesicle recycling such as COG3, CABP5, CYH1, and FRPD1, was increased. Strikingly, vision-related proteins essential for photoreceptor maintenance and retinal signaling, including RGR, RDH5, RPE65, IMPG2, and KCNJ13, were uniformly downregulated (Supplementary Figure S1). Together, these findings demonstrate that proteins associated with chromatin, mitochondrial metabolism, intracellular transport, and vision are altered early in the retina of 3xTg-AD mice.
Coupled mitochondrial and nuclear protein alterations in the AD retina
Building on the pathway-level dysregulation identified in early AD retinas, we next examined the subcellular localization of differentially expressed proteins. Cellular component enrichment analysis demonstrated that significantly altered proteins in 3xTg-AD retinas were predominantly localized to the nucleosome core, mitochondria, and chromosome (Figure 4A; Supplementary Table S2). Multiple mitochondrial proteins involved in metabolism and energy production, including LACB2, LPIN1, SRAC1, NDUB2, ARGI1, SUCB2, and ADT4, were downregulated in the AD retina (Figure 4B). In contrast, proteins associated with nuclear processes were upregulated, including histone variants and factors involved in DNA replication (RFC5), RNA processing (RNH2A, DDX55), and chromatin remodeling (SMRCD).
FIGURE 4
To determine whether the significantly altered retinal proteins identified in 3xTg-AD mice have known associations with AD in humans, we queried the Agora AD Knowledge Portal to identify proteins linked to AD pathogenesis and to explore potentially shared pathways. We found that twelve of the significantly altered retinal proteins involved in synaptic function and metabolism have prior evidence linked to AD pathogenesis (Figure 4C). In particular, several mitochondrial proteins which are changed in the 3xTG-AD retina (e.g., ADT4, SRAC1, ARGI2) were also altered in AD patients. The reduction of multiple mitochondrial proteins suggests that mitochondrial dysfunction contributes to early retinal stress in AD (Figure 4D). In particular, decreased expression of SUCB2, the β-subunit of mitochondrial succinyl-CoA ligase, indicates disruption of TCA cycle-linked metabolism and of succinyl-CoA availability, a key donor for lysine succinylation and lactylation (Weinert et al., 2013; ; ). This retrograde mitochondria-to-nucleus signaling is known to influence nuclear gene expression (; ; Tsusaka et al., 2025). Together, these findings indicate that mitochondrial dysfunction and associated nuclear responses represent an early molecular feature of retinal deficit in AD.
Early proteomic alterations in 3xTg-AD brain
Multivariate PCA demonstrated a clear separation between 3xTg-AD and WT brain proteomes, indicating substantial global differences in protein expression (Figure 5A). Differential expression analysis identified 130 significantly altered proteins, including 38 downregulated and 92 upregulated in the AD brain (Figure 5B; Supplementary Table S3). Similar to the retina, hemoglobin subunits (HBB1, HBB2, and A8DUK4) were among the most prominently altered proteins. GO analysis of biological processes showed significant enrichment of pathways related to molecular transport and the mitochondrial respiratory chain in the 3xTg-AD brain (Figure 5C; Supplementary Table S4). Consistent with this, cellular component analysis indicated that differentially expressed proteins were primarily localized to mitochondria, membrane, cytoplasmic vesicles, and synaptic compartments (Figure 5D; Supplementary Table S4). Assessment of mitochondrial pathways showed broad changes in proteins involved in oxidative phosphorylation and electron transport chains, with strong enrichment of subunits from Complex I and Complex III (Figure 5E). These included multiple NADH dehydrogenase subunits, such as NDUS1, NDUS5, NDUA6, as well as cytochrome c oxidase–associated proteins, including COX5A, CX6A1, and CX7A2.
FIGURE 5
In addition to respiratory chain alterations, proteins involved in mitochondrial protein homeostasis and autophagy were also differentially expressed (Figure 5F). These included downregulation of ribosomal and mitochondrial-associated proteins such as large ribosomal subunit protein mL48 (RM48), uL2m (RM02), uL14 m (RM14), and prohibitin 1 (PHB), as well as upregulation of autophagy-related proteins including ATG13 and TECPR2. Protein-protein interaction network analysis further highlighted clustering of altered proteins within pathways related to oxidative phosphorylation, mitophagy, and synaptic vesicle maturation (Supplementary Figure S2). To assess the relevance of these proteomic changes to AD, we examined whether altered brain proteins have prior evidence of involvement in AD. Several proteins with established links to Aβ clearance, tau regulation, and neuroinflammation were altered in 3xTg-AD brain proteome, including downregulation of the Aβ-degrading enzyme IDE and the tau deacetylase HDAC6, and upregulation of the outer mitochondrial membrane proteins VDAC1 and VDAC2, the NF-κB kinase IKKβ, and the autophagy initiator ATG13 (see Table 2). Taken together, the proteomic alterations in the 3xTg-AD brain are dominated by changes in mitochondrial proteins including proteins with established relevance to AD pathogenesis.
TABLE 2
| Protein ID | Full name | Log2FC | Localization | References |
|---|---|---|---|---|
| IDE | Insulin-degrading enzyme | −1.20 | Cytoplasm | , |
| TM40 L | Mitochondrial import receptor subunit TOM40B | 2.58 | MOM | , |
| VDAC1 | Voltage-dependent anion-selective channel protein 1 | 1.12 | MOM | Yoo et al. (2001), , Yang et al. (2024) |
| VDAC2 | Voltage-dependent anion-selective channel protein 2 | 1.05 | MOM | Yoo et al. (2001), |
| HDAC6 | Histone deacetylase 6 | −1.21 | Cytoplasm, Nucleus | |
| IKKβ | Inhibitor of nuclear factor kappa-B kinase subunit beta | 1.93 | Cytoplasm | |
| ATG13 | Autophagy-related protein 13 | 2.67 | Cytoplasm | Uddin et al. (2018), Zhang et al. (2021) |
AD-relevant proteins altered in the brain proteome of 3xTg-AD mice at 4-weeks of age. Brain proteomic analysis of 3xTg-AD mice identified several proteins with established relevance to AD pathogenesis that were significantly altered at this early stage, prior to detectable amyloid or tau pathology. |log2FC| ≥ 1; p ≤ 0.05. Mitochondria outer membrane (MOM).
Comparison of early proteomic alterations in retina and brain in 3xTg-AD mice
To investigate how early proteomic changes were shared between the retina and brain, we compared proteins significantly altered in each tissue of 3xTg-AD mice relative to WT controls. Among the 92 proteins altered in the retina and the 130 in the brain, eight proteins were significantly changed in both tissues (Figure 6A). These shared proteins included three hemoglobin-related proteins (HBB1, HBB2, and A8DUK4), and several proteins involved in metabolic regulation and protein turnover, including insulin-degrading enzyme (IDE), lactoylglutathione lyase (LGUL), E3 ubiquitin-protein ligase (PP1RB), TBC1 domain family member 23 (TBC23), and zinc finger CCCH-type containing protein 7B (F8VPP8). Except for A8DUK4 and F8VPP8, all overlapping proteins were reduced in both tissues (Figure 6B).
FIGURE 6
Pathway and cellular localization analyses further revealed common changes across the two tissues. GO analysis showed that metabolism, transport, and signaling were among the most highly enriched biological processes in both retina and brain (Figure 6C). Analysis of subcellular localization demonstrated that significantly altered proteins in both tissues were similarly distributed in cytosolic and membrane-bound compartments, while tissue-specific differences were observed in mitochondrial and nuclear fractions: mitochondrial proteins were more prominently altered in the brain (35% vs. 12% in retina), and nuclear proteins were more enriched in the retina (21% vs. 10% in brain), suggesting compartment-specific vulnerabilities in each tissue (Figure 6D). Together, these comparative analyses indicate that the retina and brain share early AD-associated proteomic alterations while also having tissue-specific changes in the initial stages of disease.
Discussion
In this study, we have identified early proteomic signatures in the retina and brain of 3xTg-AD mice. Although only eight proteins overlapped between the two tissues, both showed similar alterations in mitochondrial metabolism, oxygen, and intracellular transport. In addition to common changes, the retinal proteome showed specific alterations in vision-related, transporter, and chromatin-associated proteins, whereas the brain proteome featured changes in mitochondrial respiratory chain components and protein homeostasis. Many of these dysregulated pathways are recognized as progressive drivers or early features of AD in patients (; ; ; ; ). Our findings further support the retina as a sensitive “window to the brain” for detecting preclinical AD pathology.
Among the shared alterations, β-globin subunits HBB1, HBB2, and A8DUK4 stand out as consistently dysregulated in both tissues. While hemoglobin is classically associated with erythrocytes, neurons and glial cells also express hemoglobin to facilitate oxygen uptake under hypoxic stress (; ). Altered hemoglobin expression and localization have been reported in AD brain, including enrichment near Aβ plaques and cerebral amyloid angiopathy (Wu et al., 2004; ), as well as paradoxical depletion in neurons containing neurofibrillary tangles (; ). Beyond oxygen transport, neuronal hemoglobin has been implicated in mitochondrial function, mitochondria-to-nucleus signaling, epigenetic regulation, and autophagy (; ; ). In addition, hemoglobin can localize to the inner mitochondrial membrane (), interact with ATP synthase (), enhance mitochondrial respiration, and improve neuronal oxygen utilization in dopaminergic neurons (; ). In contrast to the brain, the role of hemoglobin in the neural retina remains unclear. Human retinal pigment epithelial (RPE) cells can synthesize and secrete hemoglobin (). In addition, hemoglobin expression in retinal macroglia and ganglion cells increases under hypoxic conditions to support cell survival (). Given the high metabolic demand and relatively low physiological oxygen availability in the retina (), locally expressed hemoglobin may help support the mitochondrial function. Thus, altered expression of hemoglobin subunits in both retina and brain at this early stage of AD may reflect a compensatory adaptation to altered mitochondrial function. However, how neuronal hemoglobins contribute to AD pathogenesis requires further investigation.
The detection of multiple proteins with established relevance to AD pathology supports the sensitivity of our approach and confirms that the 3xTg-AD model recapitulates early molecular features of the disease (Table 2). Among them, IDE, HDAC6, VDAC1/2, and ATG13 were significantly changed in 3xTg-AD brain proteome. IDE is the primary cytosolic protease responsible for Aβ degradation; its downregulation reduces Aβ clearance and is thought to contribute to amyloid accumulation prior to plaque formation. Reduced IDE expression has been reported in AD brain, particularly in the hippocampus of APOE-ε4 carriers (; ; ). HDAC6, a cytosolic deacetylase, is linked to tau modification and AD pathology. Both genetic loss and pharmacological inhibition of HDAC6 are associated with tau aggregation and altered autophagic clearance in AD models (; ; ). VDAC1 and VDAC2 are outer mitochondrial membrane channels known to interact with Aβ. VDAC1, the predominant isoform, is increased in AD brain and APP-transgenic models, where its abnormal interaction with Aβ and phosphorylated tau has been linked to impaired metabolite exchange, mitochondrial dysfunction, and apoptosis (). VDAC2 is expressed at lower levels in neurons and plays a modest, region-dependent role in AD than VDAC1 (Yoo et al., 2001; ). ATG13 is a core component of the autophagy initiation complex. Accumulating evidence has shown that impaired autophagy contributes to AD pathogenesis (Yu et al., 2005; Uddin et al., 2018; Zhang et al., 2021). However, whether changes in ATG13 reflect a compensatory response or an early maladaptive shift remains unclear.
Mitochondrial dysfunction is a key contributor in AD pathogenesis, with extensive evidence of impaired oxidative metabolism, reduced ATP production, altered tricarboxylic acid (TCA) cycle, and respiratory chain activity (; ; Wang et al., 2020; 2024). Key metabolic enzymes SUCB2 and ADT4 were reduced more than 10- and 11-fold, respectively, in the 3xTg-AD retina (Figure 4B). SUCB2, encoded by SUCLG2, catalyzes the reversible conversion of succinyl-CoA to succinate coupled with GTP synthesis (). This reaction is particularly important in the retina, where GTP is required for cGMP production during phototransduction (). Beyond its metabolic role, SUCB2 is important for protein succinylation and lactylation, linking metabolism to post-translational modifications (Figure 4D) (Zhang et al., 2023; ). Altered succinylation of mitochondrial proteins, APP, and tau has been reported in human AD brains and mouse models (; Yang et al., 2022; Zhang et al., 2025). Further studies will be required to examine if these modifications are altered in the retinal proteome at this stage. Similarly, ADT4 (SLC25A31), a mitochondrial ADP/ATP translocase responsible for exchanging matrix ATP with cytosolic ADP across the inner mitochondrial membrane, is massively reduced. Such impaired ADP/ATP exchange can compromise oxidative phosphorylation, alter mitochondrial membrane potential, and increase susceptibility to cellular stress (; ; ). The substantial reduction of SUCB2 and ADT4 in early AD retina could limit the availability of GTP and ATP, compromising mitochondrial metabolism and potentially affecting phototransduction and metabolite-driven signaling. It warrants future investigation whether these metabolic defects contribute to the downregulation of retinal proteins involved in the visual cycle, ion transport, and photoreceptor maintenance before visual impairment in 3xTg-AD mice and AD patients (; ). These findings further support that mitochondrial dysfunction emerges early in the retina in AD pathogenesis.
This study has several limitations. First, we examined a single mouse model at one early time point, which provides a snapshot of early proteomic changes rather than a longitudinal trajectory. Future studies tracking these alterations at multiple time points will be important for understanding how the proteomic alterations evolve with AD progression. Second, the 3xTg-AD model expresses human APPSwe and tauP301L under a Thy1-based neuronal promoter, which drives early neuronal overexpression of these transgenes at non-physiological levels (; ). Thus, some retinal and brain proteomic alterations may reflect transgene-driven neuronal responses rather than fully recapitulating the molecular changes of sporadic AD. Because Thy1 activity in the retina is enriched in inner retinal neurons, the whole-retina proteome likely underrepresents cell types with little or no transgene expression (; ). Third, whole-brain analysis averages proteomic signals across multiple brain regions, which may obscure region-specific vulnerability. Future studies using cell type- and region-resolved proteomics will be needed to provide greater spatial resolution of these changes. Finally, the identified candidate proteins and pathways require further biochemical and functional validation to establish their significance in AD diagnosis and pathogenesis. Despite these limitations, our work establishes that the retina and brain share overlapping yet distinct early proteomic changes, underscoring the importance of multi-tissue omics approaches for capturing the early molecular landscape of AD.
Statements
Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.
Ethics statement
The animal study was approved by Institutional Animal Care and Use Committee (IACUC) of West Virginia University. The study was conducted in accordance with the local legislation and institutional requirements.
Author contributions
AP: Data curation, Formal Analysis, Methodology, Resources, Software, Validation, Visualization, Writing – original draft, Writing – review and editing. RM: Data curation, Formal Analysis, Investigation, Methodology, Resources, Validation, Visualization, Writing – original draft, Writing – review and editing. RX: Data curation, Investigation, Methodology, Resources, Validation, Writing – review and editing. SZ: Investigation, Resources, Writing – review and editing. DH: Writing – original draft, Writing – review and editing. JD: Conceptualization, Data curation, Funding acquisition, Investigation, Project administration, Resources, Supervision, Writing – original draft, Writing – review and editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. NIH Grants (EY026030, EY031324, EY032462), the Retina Research Foundation, NIH/NIGMS grant R24GM137786 to IDeA National Resource for Quantitative Proteomics, NIH NIGMS P20GM144230 Visual Sciences COBRE grant to WVU, and an unrestricted challenge grant from Research to Prevent Blindness (RPB) to the Ophthalmology department at WVU.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcell.2026.1827348/full#supplementary-material
Abbreviations
AD, Alzheimer’s disease; WT, wild-type; 3x-Tg-AD, triple-transgenic-AD; SUCB2, succinate-CoA ligase subunit β; HBB, hemoglobin subunit β; ER, endoplasmic reticulum; Aβ, amyloid-β; GO, Gene Ontology.
References
1
AltinozM. A.GuloksuzS.Schmidt-KastnerR.KenisG.InceB.RuttenB. P. F. (2019). Involvement of hemoglobins in the pathophysiology of Alzheimer’s disease. Exp. Gerontol.126, 110680. 10.1016/j.exger.2019.110680
2
AnnaertW.De StrooperB. (2002). A cell biological perspective on Alzheimer’s disease. Annu. Rev. Cell Dev. Biol.18, 25–51. 10.1146/annurev.cellbio.18.020402.142302
3
AsanadS.FelixC. M.FantiniM.HarringtonM. G.SadunA. A.KaranjiaR. (2021). Retinal ganglion cell dysfunction in preclinical Alzheimer’s disease: an electrophysiologic biomarker signature. Sci. Rep.11 (1), 6344. 10.1038/s41598-021-85010-1
4
AshrafA. A.DaniM.SoP.-W. (2020). Low cerebrospinal fluid levels of hemopexin are associated with increased alzheimer’s pathology, hippocampal hypometabolism, and cognitive decline. Front. Mol. Biosci.7, 590979. 10.3389/fmolb.2020.590979
5
AtlanteA.ValentiD.LatinaV.AmadoroG. (2022). Dysfunction of mitochondria in alzheimer’s disease: ANT and VDAC interact with toxic proteins and aid to determine the fate of brain cells. Int. J. Mol. Sci.23 (14), 7722. 10.3390/ijms23147722
6
BayerA. U.FerrariF.ErbC. (2002). High occurrence rate of glaucoma among patients with alzheimer’s disease. Eur. Neurol.47 (3), 165–168. 10.1159/000047976
7
BelfioreR.RodinA.FerreiraE.VelazquezR.BrancaC.CaccamoA.et al (2019). Temporal and regional progression of Alzheimer’s disease-like pathology in 3xTg-AD mice. Aging Cell18 (1), e12873. 10.1111/acel.12873
8
BiagioliM.PintoM.CesselliD.ZaninelloM.LazarevicD.RoncagliaP.et al (2009). Unexpected expression of alpha- and beta-globin in mesencephalic dopaminergic neurons and glial cells. Proc. Natl. Acad. Sci. U. S. A. 106 (36), 15454–15459. 10.1073/pnas.0813216106
9
BlanksJ. C.TorigoeY.HintonD. R.BlanksR. H. (1996). Retinal pathology in Alzheimer’s disease. I. Ganglion cell loss in foveal/parafoveal retina. Neurobiol. Aging17 (3), 377–384. 10.1016/0197-4580(96)00010-3
10
BolstadB. M.IrizarryR. A.AstrandM.SpeedT. P. (2003). A comparison of normalization methods for high density oligonucleotide array data based on variance and bias. Bioinformatics19 (2), 185–193. 10.1093/bioinformatics/19.2.185
11
BrownN.AlkhayerK.ClementsR.SinghalN.GregoryR.AzzamS.et al (2016). Neuronal hemoglobin expression and its relevance to multiple sclerosis neuropathology. J. Molecular Neuroscience MN59 (1), 1–17. 10.1007/s12031-015-0711-6
12
BuchmanA. S.YuL.WilsonR. S.LimA.DaweR. J.GaiteriC.et al (2019). Physical activity, common brain pathologies, and cognition in community-dwelling older adults. Neurology92 (8), e811–e822. 10.1212/WNL.0000000000006954
13
CaroniP. (1997). Overexpression of growth-associated proteins in the neurons of adult transgenic mice. J. Neurosci. Methods71 (1), 3–9. 10.1016/s0165-0270(96)00121-5
14
Cerquera-JaramilloM. A.Nava-MesaM. O.González-ReyesR. E.Tellez-ContiC.de-la-TorreA. (2018). Visual features in alzheimer’s disease: from basic mechanisms to clinical overview. Neural Plast.2018, 2941783. 10.1155/2018/2941783
15
ChawadeA.AlexanderssonE.LevanderF. (2014). Normalyzer: a tool for rapid evaluation of normalization methods for omics data sets. J. Proteome Res.13 (6), 3114–3120. 10.1021/pr401264n
16
ChenY.-C.ChangS. C.LeeY. S.HoW. M.HuangY. H.WuY. Y.et al (2023). TOMM40 genetic variants cause neuroinflammation in alzheimer’s disease. Int. J. Mol. Sci.24 (4), 4085. 10.3390/ijms24044085
17
ChiuK.ChanT. F.WuA.LeungI. Y. P.SoK. F.ChangR. C. C. (2012). Neurodegeneration of the retina in mouse models of Alzheimer’s disease: what can we learn from the retina?Age34 (3), 633–649. 10.1007/s11357-011-9260-2
18
ChowR. L.LangR. A. (2001). Early eye development in vertebrates. Annu. Rev. Cell Dev. Biol.17 (1), 255–296. 10.1146/annurev.cellbio.17.1.255
19
ClémençonB.BabotM.TrézéguetV. (2013). The mitochondrial ADP/ATP carrier (SLC25 family): pathological implications of its dysfunction. Mol. Aspects Med.34 (2–3), 485–493. 10.1016/j.mam.2012.05.006
20
CodrichM.BertuzziM.RussoR.FrancescattoM.EspinozaS.ZentilinL.et al (2017). Neuronal hemoglobin affects dopaminergic cells’ response to stress. Cell Death & Dis.8 (1), e2538. 10.1038/cddis.2016.458
21
CookD. G.LeverenzJ. B.McMillanP. J.KulstadJ. J.EricksenS.RothR. A.et al (2003). Reduced hippocampal insulin-degrading enzyme in late-onset Alzheimer’s disease is associated with the apolipoprotein E-epsilon4 allele. Am. J. Pathology162 (1), 313–319. 10.1016/s0002-9440(10)63822-9
22
CountryM. W. (2017). Retinal metabolism: a comparative look at energetics in the retina. Brain Res.1672, 50–57. 10.1016/j.brainres.2017.07.025
23
CunhaJ. P.ProençaR.Dias-SantosA.AlmeidaR.ÁguasH.AlvesM.et al (2017). In Alzheimer’s disease: thinning of the RNFL and superior hemiretina. Graefe’s Archive Clin. Exp. Ophthalmol.255 (9), 1827–1835. 10.1007/s00417-017-3715-9
24
DeckerH.LoK. Y.UngerS. M.FerreiraS. T.SilvermanM. A. (2010). Amyloid-beta peptide oligomers disrupt axonal transport through an NMDA receptor-dependent mechanism that is mediated by glycogen synthase kinase 3beta in primary cultured hippocampal neurons. J. Neurosci. Official J. Soc. Neurosci.30 (27), 9166–9171. 10.1523/JNEUROSCI.1074-10.2010
25
Di FilippoM.ChiasseriniD.TozziA.PicconiB.CalabresiP. (2010). Mitochondria and the link between neuroinflammation and neurodegeneration. J. Alzheimer’s Disease JAD20 (Suppl. 2), S369–S379. 10.3233/JAD-2010-100543
26
DuJ.RountreeA.CleghornW. M.ContrerasL.LindsayK. J.SadilekM.et al (2016). Phototransduction influences metabolic flux and nucleotide metabolism in mouse retina. J. Biol. Chem.291 (9), 4698–4710. 10.1074/jbc.M115.698985
27
DuboisB.FeldmanH. H.JacovaC.HampelH.MolinuevoJ. L.BlennowK.et al (2014). Advancing research diagnostic criteria for Alzheimer’s disease: the IWG-2 criteria. Lancet. Neurology13 (6), 614–629. 10.1016/S1474-4422(14)70090-0
28
DumontM.HoD. J.CalingasanN. Y.XuH.GibsonG.BealM. F. (2009). Mitochondrial dihydrolipoyl succinyltransferase deficiency accelerates amyloid pathology and memory deficit in a transgenic mouse model of amyloid deposition. Free Radic. Biol. Med.47 (7), 1019–1027. 10.1016/j.freeradbiomed.2009.07.008
29
FerrerI.GómezA.CarmonaM.HuesaG.PortaS.Riera-CodinaM.et al (2011). Neuronal hemoglobin is reduced in Alzheimer’s disease, argyrophilic grain disease, Parkinson’s disease, and dementia with lewy bodies. J. Alzheimer’s Disease JAD23 (3), 537–550. 10.3233/JAD-2010-101485
30
GadhaveK.KumarD.UverskyV. N.GiriR. (2021). A multitude of signaling pathways associated with Alzheimer’s disease and their roles in AD pathogenesis and therapy. Med. Res. Rev.41 (5), 2689–2745. 10.1002/med.21719
31
GrawS.TangJ.ZafarM. K.ByrdA. K.BoldenC.PetersonE. C.et al (2020). proteiNorm - a user-friendly tool for normalization and analysis of TMT and label-free protein quantification. ACS Omega5 (40), 25625–25633. 10.1021/acsomega.0c02564
32
GrenellA.WangY.YamM.SwarupA.DilanT. L.HauerA.et al (2019). Loss of MPC1 reprograms retinal metabolism to impair visual function. Proc. Natl. Acad. Sci. U. S. A.116 (9), 3530–3535. 10.1073/pnas.1812941116
33
Gutiérrez-AguilarM.BainesC. P. (2013). Physiological and pathological roles of mitochondrial SLC25 carriers. Biochem. J.454 (3), 371–386. 10.1042/BJ20121753
34
HannaJ.DavidL. A.TouahriY.FlemingT.ScreatonR. A.SchuurmansC. (2022). Beyond genetics: the role of metabolism in photoreceptor survival, development and repair. Front. Cell Dev. Biol.10, 887764. 10.3389/fcell.2022.887764
35
HuQ.XuJ.WangL.YuanY.LuoR.GanM.et al (2023). SUCLG2 regulates mitochondrial dysfunction through succinylation in lung adenocarcinoma. Adv. Sci.10 (35), e2303535. 10.1002/advs.202303535
36
HuberW.von HeydebreckA.SültmannH.PoustkaA.VingronM. (2002). Variance stabilization applied to microarray data calibration and to the quantification of differential expression. Bioinformatics18 (Suppl. 1), S96–S104. 10.1093/bioinformatics/18.suppl_1.s96
37
HubinE.CioffiF.RozenskiJ.van NulandN. A. J.BroersenK. (2016). Characterization of insulin-degrading enzyme-mediated cleavage of Aβ in distinct aggregation states. Biochimica Biophysica Acta1860 (6), 1281–1290. 10.1016/j.bbagen.2016.03.010
38
HurleyJ. B. (2021). Retina metabolism and metabolism in the pigmented epithelium: a busy intersection. Annu. Rev. Vis. Sci.7, 665–692. 10.1146/annurev-vision-100419-115156
39
JavonilloD. I.TranK. M.PhanJ.HingcoE.KramárE. A.da CunhaC.et al (2021). Systematic phenotyping and characterization of the 3xTg-AD mouse model of alzheimer’s disease. Front. Neurosci.15, 785276. 10.3389/fnins.2021.785276
40
JiaK.DuH. (2021). Mitochondrial permeability transition: a pore intertwines brain aging and alzheimer’s disease. Cells10 (3), 649. 10.3390/cells10030649
41
JiangJ.WangH.LiW.CaoX.LiC. (2016). Amyloid plaques in retina for diagnosis in alzheimer’s patients: a meta-analysis. Front. Aging Neurosci.8, 267. 10.3389/fnagi.2016.00267
42
KazemeiniS.Nadeem-TariqA.ShihR.RafananJ.GhaniN.VidaT. A. (2024). From plaques to pathways in alzheimer’s disease: the mitochondrial-neurovascular-metabolic hypothesis. Int. J. Mol. Sci.25 (21), 11720. 10.3390/ijms252111720
43
KimJ.-I.KangB.-H. (2019). Decreased retinal thickness in patients with Alzheimer’s disease is correlated with disease severity. PLOS ONE14, e0224180. 10.1371/journal.pone.0224180
44
KimC. K.LeeY. R.OngL.GoldM.KalaliA.SarkarJ. (2022). Alzheimer’s disease: key insights from two decades of clinical trial failures. J. Alzheimer’s Disease JAD87 (1), 83–100. 10.3233/JAD-215699
45
KoronyoY.RentsendorjA.MirzaeiN.RegisG. C.SheynJ.ShiH.et al (2023). Retinal pathological features and proteome signatures of Alzheimer’s disease. Acta Neuropathol.145 (4), 409–438. 10.1007/s00401-023-02548-2
46
Koronyo-HamaouiM.KoronyoY.LjubimovA. V.MillerC. A.KoM. K.BlackK. L.et al (2011). Identification of amyloid plaques in retinas from Alzheimer’s patients and noninvasive in vivo optical imaging of retinal plaques in a mouse model. NeuroImage54, S204–S217. 10.1016/j.neuroimage.2010.06.020
47
KremenW. S.PanizzonM. S.ElmanJ. A.GranholmE. L.AndreassenO. A.DaleA. M.et al (2019). Pupillary dilation responses as a midlife indicator of risk for Alzheimer’s disease: association with Alzheimer’s disease polygenic risk. Neurobiol. Aging83, 114–121. 10.1016/j.neurobiolaging.2019.09.001
48
KundraR.CiryamP.MorimotoR. I.DobsonC. M.VendruscoloM. (2017). Protein homeostasis of a metastable subproteome associated with Alzheimer’s disease. Proc. Natl. Acad. Sci. U. S. A.114 (28), E5703–E5711. 10.1073/pnas.1618417114
49
KusneY.WolfA. B.TownleyK.ConwayM.PeymanG. A. (2017). Visual system manifestations of Alzheimer’s disease. Acta Ophthalmol.95 (8), e668–e676. 10.1111/aos.13319
50
La MorgiaC.Ross-CisnerosF. N.KoronyoY.HannibalJ.GallassiR.CantalupoG.et al (2016). Melanopsin retinal ganglion cell loss in A lzheimer disease. Ann. Neurology79 (1), 90–109. 10.1002/ana.24548
51
LaneC. A.HardyJ.SchottJ. M. (2018). Alzheimer’s disease. Eur. J. Neurology25 (1), 59–70. 10.1111/ene.13439
52
LiuR.RenX.ParkY. E.FengH.ShengX.SongX.et al (2025). Nuclear GTPSCS functions as a lactyl-CoA synthetase to promote histone lactylation and gliomagenesis. Cell Metab.37 (2), 377–394. 10.1016/j.cmet.2024.11.005
53
LondonA.BenharI.SchwartzM. (2013). The retina as a window to the brain-from eye research to CNS disorders. Nat. Rev. Neurol.9 (1), 44–53. 10.1038/nrneurol.2012.227
54
OddoS.CaccamoA.ShepherdJ. D.MurphyM. P.GoldeT. E.KayedR.et al (2003). Triple-transgenic model of Alzheimer’s disease with plaques and tangles: intracellular abeta and synaptic dysfunction. Neuron39 (3), 409–421. 10.1016/s0896-6273(03)00434-3
55
On behalf of the FACEHBI study group (2020). Association between retinal thickness and β-amyloid brain accumulation in individuals with subjective cognitive decline: fundació ACE healthy brain initiative. Alzheimer’s Res. & Ther.12 (1), 37. 10.1186/s13195-020-00602-9
56
OnyangoI. G.DennisJ.KhanS. M. (2016). Mitochondrial dysfunction in alzheimer’s disease and the rationale for bioenergetics based therapies. Aging Dis.7 (2), 201–214. 10.14336/AD.2015.1007
57
OttoyJ.OzzoudeM.ZukotynskiK.AdamoS.ScottC.GaudetV.et al (2023). Vascular burden and cognition: mediating roles of neurodegeneration and amyloid PET. Alzheimer’s & Dementia J. Alzheimer’s Assoc.19 (4), 1503–1517. 10.1002/alz.12750
58
PanR.-Y.HeL.ZhangJ.LiuX.LiaoY.GaoJ.et al (2022). Positive feedback regulation of microglial glucose metabolism by histone H4 lysine 12 lactylation in Alzheimer’s disease. Cell Metab.34 (4), 634–648. 10.1016/j.cmet.2022.02.013
59
PeriasamyA.MitchellN.ZaytsevaO.ChahalA. S.ZhaoJ.ColmanP. M.et al (2022). An increase in mitochondrial TOM activates apoptosis to drive retinal neurodegeneration. Sci. Rep.12 (1), 21634. 10.1038/s41598-022-23280-z
60
PerlD. P. (2010). Neuropathology of Alzheimer’s disease. Mt. Sinai J. Med. N. Y.77 (1), 32–42. 10.1002/msj.20157
61
PhillipsD.AponteA. M.FrenchS. A.ChessD. J.BalabanR. S. (2009). Succinyl-CoA synthetase is a phosphate target for the activation of mitochondrial metabolism. Biochemistry48 (30), 7140–7149. 10.1021/bi900725c
62
PillaiJ. A.BermelR.Bonner-JacksonA.Rae-GrantA.FernandezH.BenaJ.et al (2016). Retinal nerve fiber layer thinning in alzheimer’s disease: a case-control study in comparison to normal aging, parkinson’s disease, and non-alzheimer’s dementia. Am. J. Alzheimer’s Dis. Other Dementias31 (5), 430–436. 10.1177/1533317515628053
63
QureshiT.ChinnathambiS. (2022). Histone deacetylase-6 modulates tau function in Alzheimer’s disease. Biochimica Biophysica Acta. Mol. Cell Res.1869 (8), 119275. 10.1016/j.bbamcr.2022.119275
64
RajanK. B.WeuveJ.BarnesL. L.McAninchE. A.WilsonR. S.EvansD. A. (2021). Population estimate of people with clinical Alzheimer’s disease and mild cognitive impairment in the United States (2020–2060). Alzheimer’s & Dementia17 (12), 1966–1975. 10.1002/alz.12362
65
RatanY.RajputA.MaleysmS.PareekA.JainV.PareekA.et al (2023). An insight into cellular and molecular mechanisms underlying the pathogenesis of neurodegeneration in alzheimer’s disease. Biomedicines11 (5), 1398. 10.3390/biomedicines11051398
66
RisacherS. L.WuDunnD.TallmanE. F.WestJ. D.GaoS.FarlowM. R.et al (2020). Visual contrast sensitivity is associated with the presence of cerebral amyloid and tau deposition. Brain Commun.2 (1), fcaa019. 10.1093/braincomms/fcaa019
67
RitchieM. E.PhipsonB.WuD.HuY.LawC. W.ShiW.et al (2015). Limma powers differential expression analyses for RNA-Sequencing and microarray studies. Nucleic Acids Res.43 (7), e47. 10.1093/nar/gkv007
68
RowlandH. A.MoxonS. R.CorbettN. J.HansonK.FisherK.KellettK. A. B.et al (2023). Inhibition of insulin-degrading enzyme in human neurons promotes amyloid-β deposition. Neuronal Signal.7 (4), NS20230016. 10.1042/NS20230016
69
Salobrar-GarcíaE.Rodrigues-NevesA. C.RamírezA. I.de HozR.Fernández-AlbarralJ. A.López-CuencaI.et al (2020). Microglial activation in the retina of a triple-transgenic alzheimer’s disease mouse model (3xTg-AD). Int. J. Mol. Sci.21 (3), 816. 10.3390/ijms21030816
70
SchelshornD. W.SchneiderA.KuschinskyW.WeberD.KrügerC.DittgenT.et al (2009). Expression of hemoglobin in rodent neurons. J. Cereb. Blood Flow Metabolism29 (3), 585–595. 10.1038/jcbfm.2008.152
71
SchnöderL.QuanW.YuY.TomicI.LuoQ.HaoW.et al (2023). Deficiency of IKKβ in neurons ameliorates Alzheimer’s disease pathology in APP ‐ and tau‐transgenic mice. FASEB J.37 (2), e22778. 10.1096/fj.202201512R
72
SearleB. C.PinoL. K.EgertsonJ. D.TingY. S.LawrenceR. T.MacLeanB. X.et al (2018). Chromatogram libraries improve peptide detection and quantification by data independent acquisition mass spectrometry. Nat. Commun.9 (1), 5128. 10.1038/s41467-018-07454-w
73
SelkoeD. J. (2000). Toward a comprehensive theory for Alzheimer’s disease. Hypothesis: alzheimer’s disease is caused by the cerebral accumulation and cytotoxicity of amyloid beta-protein. Ann. N. Y. Acad. Sci.924, 17–25. 10.1111/j.1749-6632.2000.tb05554.x
74
ShephardF.Greville-HeygateO.MarshO.AndersonS.ChakrabartiL. (2014). A mitochondrial location for haemoglobins--dynamic distribution in ageing and Parkinson’s disease. Mitochondrion14 (1), 64–72. 10.1016/j.mito.2013.12.001
75
Shoshan-BarmatzV.Nahon-CrystalE.Shteinfer-KuzmineA.GuptaR. (2018). VDAC1, mitochondrial dysfunction, and Alzheimer’s disease. Pharmacol. Res.131, 87–101. 10.1016/j.phrs.2018.03.010
76
SongA.JohnsonN.AyalaA.ThompsonA. C. (2021). Optical coherence tomography in patients with alzheimer’s disease: what can it tell us?Eye Brain13, 1–20. 10.2147/EB.S235238
77
SunL.ZhouR.YangG.ShiY. (2017). Analysis of 138 pathogenic mutations in presenilin-1 on the in vitro production of Aβ42 and Aβ40 peptides by γ-secretase. Proc. Natl. Acad. Sci. U. S. A.114 (4), E476–E485. 10.1073/pnas.1618657114
78
SuzukiN.CheungT. T.CaiX. D.OdakaA.OtvosL.EckmanC.et al (1994). An increased percentage of long amyloid beta protein secreted by familial amyloid beta protein precursor (beta APP717) mutants. Science264 (5163), 1336–1340. 10.1126/science.8191290
79
TezelT. H.GengL.LatoE. B.SchaalS.LiuY.DeanD.et al (2009). Synthesis and secretion of hemoglobin by retinal pigment epithelium. Investigative Ophthalmol. & Vis. Sci.50 (4), 1911–1919. 10.1167/iovs.07-1372
80
TezelG.YangX.LuoC.CaiJ.KainA. D.PowellD. W.et al (2010). Hemoglobin expression and regulation in glaucoma: insights into retinal ganglion cell oxygenation. Investigative Ophthalmol. & Vis. Sci.51 (2), 907–919. 10.1167/iovs.09-4014
81
TianZ.JiX.LiuJ. (2022). Neuroinflammation in vascular cognitive impairment and dementia: current evidence, advances, and prospects. Int. J. Mol. Sci.23 (11), 6224. 10.3390/ijms23116224
82
TrzeciakiewiczH.AjitD.TsengJ. H.ChenY.AjitA.TabassumZ.et al (2020). An HDAC6-dependent surveillance mechanism suppresses tau-mediated neurodegeneration and cognitive decline. Nat. Commun.11 (1), 5522. 10.1038/s41467-020-19317-4
83
TsusakaT.NajarM. A.SharmaI.MarcinkiewiczM. M.CrispimC. V. D. S.SnyderN. W.et al (2025). Class I histone deacetylases catalyze lysine lactylation. BioRxiv Prepr. Serv. Biol., 2025. 10.1101/2025.02.25.640220
84
UddinM. S.StachowiakA.MamunA. A.TzvetkovN. T.TakedaS.AtanasovA. G.et al (2018). Autophagy and alzheimer’s disease: from molecular mechanisms to therapeutic implications. Front. Aging Neurosci.10, 04. 10.3389/fnagi.2018.00004
85
WangW.ZhaoF.MaX.PerryG.ZhuX. (2020). Mitochondria dysfunction in the pathogenesis of Alzheimer’s disease: recent advances. Mol. Neurodegener.15 (1), 30. 10.1186/s13024-020-00376-6
86
WangS.LiaoZ.ZhangQ.HanX.LiuC.WangJ. (2024). Mitochondrial dysfunction in Alzheimer’s disease: a key frontier for future targeted therapies. Front. Immunol.15, 1484373. 10.3389/fimmu.2024.1484373
87
WeinertB. T.SchölzC.WagnerS. A.IesmantaviciusV.SuD.DanielJ. A.et al (2013). Lysine succinylation is a frequently occurring modification in prokaryotes and eukaryotes and extensively overlaps with acetylation. Cell Rep.4 (4), 842–851. 10.1016/j.celrep.2013.07.024
88
Wong-RileyM. T. T. (2010). Energy metabolism of the visual system. Eye Brain2, 99–116. 10.2147/EB.S9078
89
WuC.-W.LiaoP. C.YuL.WangS. T.ChenS. T.WuC. M.et al (2004). Hemoglobin promotes abeta oligomer formation and localizes in neurons and amyloid deposits. Neurobiol. Dis.17 (3), 367–377. 10.1016/j.nbd.2004.08.014
90
XuR.RitzB. K.WangY.HuangJ.ZhaoC.GongK.et al (2020). The retina and retinal pigment epithelium differ in nitrogen metabolism and are metabolically connected. J. Biol. Chem.295 (8), 2324–2335. 10.1074/jbc.RA119.011727
91
YaffeK.KanayaA.LindquistK.SimonsickE. M.HarrisT.ShorrR. I.et al (2004). The metabolic syndrome, inflammation, and risk of cognitive decline. JAMA292 (18), 2237–2242. 10.1001/jama.292.18.2237
92
YangY.TapiasV.AcostaD.XuH.ChenH.BhawalR.et al (2022). Altered succinylation of mitochondrial proteins, APP and tau in Alzheimer’s disease. Nat. Commun.13 (1), 159. 10.1038/s41467-021-27572-2
93
YangY.JiaX.YangX.WangJ.FangY.YingX.et al (2024). Targeting VDAC: a potential therapeutic approach for mitochondrial dysfunction in Alzheimer’s disease. Brain Res.1835, 148920. 10.1016/j.brainres.2024.148920
94
YooB. C.FountoulakisM.CairnsN.LubecG. (2001). Changes of voltage-dependent anion-selective channel proteins VDAC1 and VDAC2 brain levels in patients with Alzheimer’s disease and Down syndrome. Electrophoresis22 (1), 172–179. 10.1002/1522-2683(200101)22:1%253C172::AID-ELPS172%253E3.0
95
YuW. H.CuervoA. M.KumarA.PeterhoffC. M.SchmidtS. D.LeeJ. H.et al (2005). Macroautophagy--a novel beta-amyloid peptide-generating pathway activated in Alzheimer’s disease. J. Cell Biol.171 (1), 87–98. 10.1083/jcb.200505082
96
ZhangZ.YangX.SongY. Q.TuJ. (2021). Autophagy in Alzheimer’s disease pathogenesis: therapeutic potential and future perspectives. Ageing Res. Rev.72, 101464. 10.1016/j.arr.2021.101464
97
ZhangX.LiuJ.ChengY.ChenK.ChenY.ZhuH.et al (2023). Metabolic enzyme Suclg2 maintains tolerogenicity of regulatory dendritic cells diffDCs by suppressing lactb succinylation. J. Autoimmun.138, 103048. 10.1016/j.jaut.2023.103048
98
ZhangX.LiuY.RekowskiM. J.WangN. (2025). Lactylation of tau in human Alzheimer’s disease brains. Alzheimer’s & Dementia21 (2), e14481. 10.1002/alz.14481
99
ZhuS.YamM.WangY.LintonJ. D.GrenellA.HurleyJ. B.et al (2018). Impact of euthanasia, dissection and postmortem delay on metabolic profile in mouse retina and RPE/choroid. Exp. Eye Res.174, 113–120. 10.1016/j.exer.2018.05.032
Summary
Keywords
alzheimer’s disease, brain, mitochondria, neurodegenerative disorders, proteomics, retina
Citation
Puja A, McNeel R, Xu R, Zhu S, Hansman D and Du J (2026) Early proteomic signatures of Alzheimer’s disease in the retina and brain of 3xTg-AD mice. Front. Cell Dev. Biol. 14:1827348. doi: 10.3389/fcell.2026.1827348
Received
10 March 2026
Revised
15 April 2026
Accepted
30 April 2026
Published
18 May 2026
Volume
14 - 2026
Edited by
Daisy Y. Shu, University of New South Wales, Australia
Reviewed by
Akihiro Ikeda, University of Wisconsin-Madison, United States
Michael Landowski, Wayne State University, United States
Updates
Copyright
© 2026 Puja, McNeel, Xu, Zhu, Hansman and Du.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Jianhai Du, Jianhai.du@wisc.edu
† These authors have contributed equally to this work
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.