Abstract

   Disruptions in synaptic transmission and plasticity are early hallmarks
   of Alzheimer’s disease (AD). Endosomal trafficking, mediated by the
   retromer complex, is essential for intracellular protein sorting,
   including the regulation of amyloid precursor protein (APP) processing.
   The VPS35 subunit, a key cargo-recognition component of the retromer,
   has been implicated in neurodegenerative diseases, with mutations such
   as L625P linked to early-onset AD. Despite growing evidence for
   retromer dysfunction in AD, its role in synaptic pathology and
   neuroinflammation remains incompletely understood. Here, we investigate
   the acute molecular effects of retromer stabilization in the 5xFAD
   mouse model of AD using the pharmacological chaperones R55 and R33,
   previously identified to enhance VPS35 stability. Following
   intracranial stereotaxic injections, we performed transcriptomic
   profiling, quantitative histology, and immunohistochemistry to assess
   synaptic function, neuroinflammation, and endosomal trafficking. Our
   findings reveal that retromer stabilization reverses multiple
   AD-associated molecular changes. R55 treatment significantly reduced
   Aβ-related pathology, normalized synaptic gene expression, and restored
   long-term potentiation (LTP)-associated pathways, including Gria1 (AMPA
   receptors), Grip1, and semaphorin/plexin signaling. Additionally,
   retromer stabilization counteracted dysregulated calcium signaling by
   modulating Ryr2 and L-type calcium channel expression. Beyond synaptic
   effects, we observed broad transcriptional and structural changes in
   the endosomal system. Notably, R55 treatment decreased VPS13 family
   gene expression, implicated in membrane contact site regulation, while
   increasing RAB7 levels, suggesting enhanced late-endosomal recycling.
   VPS35-positive vesicles were redistributed away from the nucleus,
   indicating restored intracellular trafficking dynamics. In the
   neuroinflammatory domain, retromer stabilization modulated microglial
   activation, shifting towards a profile characterized by balanced
   pro-inflammatory (Il1, Nfkb2) and anti-inflammatory (Il4r, Il13ra1,
   Stat6) markers, consistent with disease-associated microglia (DAM)
   phenotypes. Together, these findings demonstrate that retromer
   dysfunction contributes to key AD pathologies, including synaptic
   dysfunction and neuroinflammation, and that pharmacological retromer
   stabilization can restore cellular homeostasis. Given that 5xFAD mice
   lack direct VPS35 mutations, our results suggest that
   retromer-targeting strategies may be applicable to both familial and
   sporadic AD, offering a promising therapeutic avenue for modifying
   disease progression.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s40478-025-02096-8.

Introduction

   Synaptic dysfunction is an early hallmark of Alzheimer's disease (AD)
   and other neurodegenerative disorders, driven by disruptions in
   synaptic transmission and plasticity. In AD, beta-amyloid (Aβ)
   accumulates within synapses, impairing neurotransmitter release,
   receptor signalling, and dendritic spine stability, strongly
   implicating Aβ in synaptic pathology [[28]7, [29]28]. However, the
   mechanisms underlying synaptic Aβ accumulation remain incompletely
   understood, and effective intervention strategies are lacking.

   The endolysosomal pathway is crucial for intracellular trafficking
   including cell membrane vesicle transport and recycling [[30]31].
   Clathrin-mediated endocytosis initiates the formation of early
   endosomes, which mature into multivesicular bodies and are subsequently
   routed either to the Golgi apparatus via the retrograde pathway or
   recycled to the plasma membrane. The retromer complex is critical in
   this process because it determines cargo selection for endosomal
   transport. The retromer complex plays a crucial role in this process by
   selecting cargo for endosomal transport, including neuronal receptors,
   but also amyloid precursor protein (APP) and APP-processing enzymes.
   Given the rapid turnover of synaptic receptors via endocytosis, any
   disruption in cargo selection can significantly impact synaptic
   function [[31]13]. Moreover, the endosomal system is integral to
   protein maturation, as proteins critical for synaptic activity,
   including APP, undergo proteolytic processing en route from the Golgi
   to the plasma membrane [[32]34, [33]37, [34]39, [35]43].

   A key component of the retromer complex is VPS35, which regulates its
   assembly, stability, and cargo recognition [[36]6]. Mutations in VPS35,
   such as D620N, are linked to autosomal dominant Parkinson’s disease
   (PARK17) [[37]42, [38]46], while the L625P variant has been associated
   with early-onset AD [[39]34]. The retromer complex influences APP
   processing by directing its trafficking within the endosomal system,
   and alterations in this pathway precede Aβ deposition in AD [[40]9].
   Notably, VPS35 expression declines early in the entorhinal cortex,
   mirroring the spatiotemporal progression of AD pathology [[41]36].
   Thus, in AD, early dysfunction in the endosomal and retromer system
   coincide with the synaptic dysfunctions. VPS35 dysfunction leads to APP
   accumulation in early endosomes, increased Aβ production and exosomal
   release, and enlarged axonal endosomes [[42]9, [43]16, [44]20].
   Furthermore, Aβ itself downregulates VPS35 protein levels, with Aβ
   colocalizing with VPS35-positive endosomes [[45]1]. Interestingly, the
   retromer complex also plays a role in microglial function, influencing
   microglial activation states that are critical in AD pathogenesis
   [[46]33].

   Given the pivotal role of VPS35, efforts have been made to
   pharmacologically stabilize the retromer complex. A small-molecule
   screening identified two thiopenthene thiourea compounds, R55 and R33,
   as pharmacological chaperones that enhance the stability of the binding
   of VPS35-VPS29 [[47]26]. In vitro, these compounds increased VPS35 and
   VPS26 protein levels in mouse primary cortical neurons and reduced APP
   processing, lowering endogenous Aβ42 and Aβ40 levels at low micromolar
   concentrations (5 µM) [[48]26]. It is now believed that the effect of
   R55 and R33 is primarily by restoring the function and localization of
   the retromer complex [[49]10]. However, despite these promising
   results, there is limited understanding of the therapeutic potential of
   retromer stabilization in in vivo AD models, exacerbated by the
   retraction of several key publications in the field [[50]22, [51]41].
   Furthermore, while in vitro data suggest Aβ involvement, the broader
   molecular mechanisms underpinning retromer-mediated neuroprotection
   remain unclear. Given that VPS35 interacts with multiple key players in
   AD pathophysiology, its dysfunction may serve as a central failure hub
   in disease progression.

   In this study, we investigate the acute molecular effects of retromer
   stabilization in the 5xFAD/C57BL6 murine AD model following 24-h
   intraparenchymal injections of R55 and R33. Our goal is to map the
   downstream signalling pathways affected by retromer stabilization and
   assess the potential to reverse AD-associated molecular events. Given
   the limited pharmacokinetic data on these compounds and the challenges
   of long-term intraparenchymal delivery, we focus on the immediate
   biochemical responses to VPS35 stabilization.

Materials and methods

Drugs

   The thiophene thioureas, R55 and R33, were obtained as hydrochloride
   salts from CaymanChem (Ann Arbor, MI) and MedKoo (Morrisville, NC),
   respectively. To prevent degradation due to hydrolysis and oxidation,
   the compounds were dissolved in high-purity anhydrous DMSO at a
   concentration of 300 mM, aliquoted, and stored at − 70 °C. Immediately
   before intracranial microinjection, one aliquot was thawed and diluted
   to 10 mM under sterile conditions using freshly prepared 20%
   sulfobutylether beta-cyclodextrin (Thermo Fisher) in saline. A vehicle
   control solution containing the same proportions of DMSO and
   beta-cyclodextrin (final concentration of 3% DMSO) was prepared
   alongside the drug solutions to ensure consistency across treatments.

   Although pharmacokinetic data for R55 and R33 is not available, the
   estimated final cortical concentration following a 500 nL injection was
   approximately 25 µM, assuming homogeneous distribution. This
   concentration exceeds the estimated dissociation constant (Kd) for
   VPS35 stabilization (~ 5 µM) and the IC₅₀ for Aβ reduction (~ 12 µM) in
   primary neurons, as previously described (Mecozzi et al., 2014).

Animals and surgeries

   Experiments were conducted using 16-week-old female 5xFAD/C57BL6 mice
   (B6), specifically Cg-Tg(APPSwFlLon, PSEN1*M146L*L286V)6799Vas/Mmjax,
   obtained from The Jackson Laboratory (JAX) and locally bred with
   C57BL/6 J mice from the same source. Compounds were administered via
   stereotaxic microinjection into the lateral parietal associative cortex
   using the following coordinates: caudal − 2.06 mm, lateral −1.5 mm, and
   ventral − 0.8 mm. Infusions were performed at a rate of 100 nL/min,
   with a total volume of 500 nL.

   Twenty-four hours post-injection, mice were euthanized, and brains were
   rapidly dissected. For the transcriptomics study, six mice per
   treatment group (n = 6 per group) were used. A cortical tissue block
   measuring approximately 1 mm3 was isolated from each hemisphere, with
   the injection site serving as the focal point in the ipsilateral
   hemisphere. The contralateral hemisphere was used as an internal
   control. Care was taken during microdissection to ensure that no parts
   of the corpus callosum were included, thereby preventing contamination
   from interhemispheric structures. To preserve RNA integrity, tissue
   blocks were submerged in RNAlater (Thermo Fisher) and stored at − 70 °C
   until processing. For histological analysis, an independent cohort of
   mice (n = 3 per treatment) underwent intracardiac perfusion with PBS
   followed by 4% PFA in PBS before whole-brain dissection. Tissue
   assignments for drug or vehicle treatment were randomized, and
   researchers were blinded to treatment identity until the completion of
   data analysis. All procedures were approved by the Animal Ethics
   Committee of Linköping University and conducted in accordance with
   ethical guidelines for animal research.

Expression profiling

   Total RNA was extracted using the RNeasy + Mini Kit (Qiagen) and
   processed with the QIAcube automated system (Qiagen). RNA integrity was
   assessed using Bioanalyzer RNA 6000 Nano Chips (Agilent), and all
   samples had an RNA integrity number (RIN) exceeding 8.5. RNA
   quantification was performed using Quant-iT (Invitrogen), and indexed
   sequencing libraries were constructed with the Illumina Stranded mRNA
   Prep Kit. Deep RNA sequencing was conducted using a NextSeq 2000
   (Illumina, P50 flow cell) following the standard Illumina pipeline.

   Paired-end reads were aligned to the mouse GRCm39 genome (June 2020)
   using HISAT2, with sequencing depths ranging between 20 and 40 million
   mapped reads per sample. Differential expression analysis was performed
   using DESeq2 within the R (R Core Team, 2023, [52]www.r-project.org)
   and Bioconductor [[53]14] framework, comparing gene expression in the
   injected hemisphere to that in the contralateral hemisphere. Additional
   comparisons were made between treatment groups, including R55 vs.
   vehicle, R33 vs. vehicle, and R55 vs. R33. Pathway and functional
   enrichment analyses were conducted using multiple databases, including
   Gene Ontology (GO), InterPro, KEGG, Monarch, Pfam, SMART, STRINGdb,
   UniProtKB, and WikiPathways.

   To contextualize transcriptional alterations observed in 5xFAD mice,
   external reference datasets were used, including publicly available
   RNA-seq data from MODEL-AD datasets ([54]GSE168137, [[55]11]), which
   document transcriptomic changes in 5xFAD mice across disease
   progression, from Bliim, Leshchyns’ka, Keable, Chen, Curry-Hyde, Gray,
   Sytnyk and Janitz [[56]5] ([57]GSE110908), which describes gene
   expression changes following chemically induced long-term potentiation
   (cLTP) and [[58]19] describing the disease-associated microglia (DAM)
   phenotypes in 5xFAD.

Quantitative histology

   Following transcardial perfusion, brains were cryoprotected in sucrose
   in 4% PFA until they lost buoyancy, then snap frozen. Sections were
   obtained using a cryostat and immunostained with the following primary
   antibodies: anti-VPS35 (clone 7E4, NBP2-78823, Novus Biologicals, CO),
   anti-RAB7 (clone E9O7, Cell Signaling, MA), anti-VPS13b (PA5-34406,
   Invitrogen), anti-SORL1 (NB100-785, Novus) and anti-IBA1 (ab5076,
   Abcam).

   Fluorescence microscopy was performed using a Zeiss LSM700 confocal
   microscope, and image deconvolution was applied for signal enhancement.
   Cortical layer V neurons adjacent to the injection site, but
   sufficiently distant from the needle track to maintain normal
   parenchymal morphology, were segmented using QuPath [[59]2].
   Fluorescence intensity was quantified in approximately 700 segmented
   neurons per protein. The quantification was done in the cortex layer V,
   aiming to analyse as close to the injection track as possible but
   without artefacts from the procedure. For the Sorl1 study, the
   morphometry, location and distance to nuclear envelope of about 20,000
   Sorl1 + particles was analysed using ImageJ DiAna plugin [[60]12].

Statistical analysis

   All data were analyzed using R, with results presented as
   mean ± standard error (SE). For multiple comparisons, FDR corrected
   p-values are presented. The statistical significance threshold was set
   at α = 0.05. Two-tailed statistical tests were applied for comparisons.

Results

Selection of the 5xFAD model for retromer stabilization studies

   To determine the most suitable mouse model for investigating retromer
   dysfunction in AD, we analysed temporal expression patterns of key
   genes within the retromer interactome and its cargo across multiple
   transgenic AD models, including APP NL-F knock-in, APP NL-G-F knock-in,
   APP/PS1, 3xTgAD, and 5xFAD (Fig. [61]1A). We compared these profiles to
   human AD datasets, assessing how gene expression evolved with disease
   progression and transgenic model age (Fig. [62]1B). Among these models,
   5xFAD exhibited the most extensive and consistent changes, closely
   mirroring human AD pathology. Based on these findings, we selected
   5xFAD mice as the primary model for this study.

Fig. 1.

   [63]Fig. 1
   [64]Open in a new tab

   Temporal evolution of retromer interactome genes and cargo in
   transgenic AD models and human AD cohorts. A Temporal progression of
   retromer-related gene expression across multiple transgenic AD models.
   Expression changes are color-coded relative to the mean expression
   across transgenic and wild-type littermates at all ages and normalized
   to the standard deviation (SD). To enhance clarity, wild-type
   expression levels are not shown. B Correlation between expression of
   equivalent genes in human AD cohorts and disease progression, assessed
   by Mini-Mental State Examination (MMSE) scores (data from
   [65]GSE185909). In transgenic mouse models, expression was analyzed as
   a function of age (data from the same GSE dataset as panel A).
   Correlations are presented separately by sex, with significant
   correlations (p < 0.05) indicated by"o"

Retromer stabilization by R55 restores synaptic and neuroinflammatory
pathways

   To assess the impact of retromer stabilization, 5xFAD mice received
   intracranial stereotaxic injections of R55 or R33, achieving an
   estimated final brain concentration of 25 µM. Vehicle-only injections
   served as controls. Twenty-four hours post-injection, brains were
   dissected for differential gene expression analysis. Hematoxylin and
   eosin staining of sections from the injection site showed the injection
   canal with intact surrounding tissue without signs of toxicity from the
   treatments. Neurons appeared with normal morphology after both vehicle,
   R55 and R33 injections (Supplementary Fig. [66]1), sections from R55
   and R33 treated animals appeared to have a slight increase in
   microglia. Out of 19,646 detected genes, 218 (1.10%) were significantly
   differentially expressed in R55-treated mice compared to vehicle
   controls (Fig. [67]2A, [68]GSE267989). R33 treatment resulted in 19
   differentially expressed genes (0.10%), with most showing changes
   similar to R55 but of lower magnitude. Direct comparison of R55 vs. R33
   revealed only 11 significantly altered genes (0.05%), supporting the
   hypothesis that R55 and R33 share a similar mechanism of action, with
   R55 exhibiting stronger effects at this time point. Thus, we focused
   mainly on R55 for the rest of the study, unless otherwise stated.
   Microinjection alone caused 171 genes (0.87%) to be differentially
   expressed between the injected and contralateral hemispheres. However,
   only four genes (Cdca5, Esyt1, Ifi30, Mmp12) overlapped with the R55
   vs. vehicle dataset, suggesting minimal confounding effects from
   injection trauma (data not shown).

Fig. 2.

   [69]Fig. 2
   [70]Open in a new tab

   Differential gene expression and pathway enrichment following retromer
   complex stabilization by R55 and R33 in 5xFAD mice. A Summary of
   differentially expressed genes (DEGs) after treatment with R55, R33, or
   vehicle (Veh) in the injected (Inj) and contralateral (Con)
   hemispheres. Treatment and hemisphere were included as separate factors
   in the statistical model, and pairwise contrasts were conducted for
   each. B Interaction network of DEGs between R55 and vehicle, with
   significantly higher connectivity than expected by chance
   (p < 1 × 10^−16). C Heatmap of significant DEGs between R55, R33 and
   vehicle across individual samples. D Pathway enrichment analysis of
   significantly altered genes in the R55 vs. vehicle comparison,
   color-coded according to the database source. Enrichment significance
   is represented as –log₁₀ (False Discovery Rate, FDR). All displayed
   terms are significant (p < 0.05). E Volcano plot illustrating log₂ fold
   changes of genes from the Alzheimer’s WP2045 dataset under different
   treatment conditions (R55, R33, and vehicle)

   Network analysis of the R55-regulated genes identified 247 interactions
   with an average node degree of 2.38 (Fig. [71]2B), significantly
   exceeding the expected 101 interactions from a random dataset
   (p = 1.10e-16), demonstrating that R55 selectively modulates
   interconnected pathways. Hierarchical clustering revealed distinct gene
   expression signatures across hemispheres and treatment groups
   (Fig. [72]2C). Interestingly, gene expression changes were observed in
   both ipsilateral and contralateral hemispheres, suggesting widespread
   parenchymal distribution of the compounds.

   Enrichment analysis using multiple databases, revealed that the
   majority of the enriched GO terms were neuronal and synaptic-related,
   as well as some related to calcium handling and axonal organization
   (Fig. [73]2D). The AD-related pathway WP2075 was significantly altered,
   along with synaptic organization pathways and other CNS-related
   phenotypes. Structural protein domain enrichment analysis highlighted
   common domains in the differential expression genes, specifically the
   ones related to PSD-95, Egf-like, and Vps13 families (Fig. [74]2E).
   Notably, a volcano plot of WP2075 genes (Fig. [75]2F) confirmed that
   both R55 and R33 modulated gene expression in the same direction, with
   R55 exerting stronger effects.

Retromer complex stabilization reverses synaptic deficits in 5xFAD mice

   To evaluate the potential neuroprotective effects of retromer-directed
   therapy, we compared R55-induced gene expression changes with data from
   the MODEL-AD 5xFAD mouse reference dataset [[76]11]. We selected
   relevant enrichment clusters and plotted gene expression changes at
   4 months (dots) and 8 months (bars) of age in red, and overlaid
   R55-induced changes in green (Fig. [77]3). As none of the databases
   contain a defined cluster for retromer complex-related genes, we
   manually curated a set of retromer genes that also included relevant
   cargos to evaluate the effects in this cluster. For most genes, R55
   treatment induced expression changes of similar magnitude but reversed
   to those observed at 8 months in 5xFAD mice, suggesting that retromer
   stabilization normalizes AD-related transcriptional alterations. This
   effect was largely specific to AD-related pathways, as genes from the
   Parkinson’s disease-associated WP2371 pathway showed minimal changes,
   with none reaching significance in the R55 vs. vehicle comparison.

Fig. 3.

   [78]Fig. 3
   [79]Open in a new tab

   Comparative pathway analysis of retromer complex stabilization in 5xFAD
   mice relative to MODEL-AD datasets. Comparison of log₂ fold changes
   between R55-treated and vehicle-treated 5xFAD mice (green) and
   age-matched 5xFAD vs. wild-type (WT) controls (red) from the MODEL-AD
   dataset. For MODEL-AD data, dots represent 4-month-old mice, and solid
   bars represent 8-month-old mice. Gene sets were curated from multiple
   databases, avoiding redundancy and overly broad categories (e.g.,
   abnormal nervous system phenotypes). Additional manually curated gene
   sets include retromer-associated genes, as no existing database
   comprehensively covered this group. Only genes significantly altered in
   the R55 vs. vehicle comparison are shown, except for WP2371
   (Parkinson’s disease pathway), where all significantly altered genes
   from both comparisons are displayed

   To examine the effect on synaptic-related proteins in more detail, we
   investigated changes in the gene set corresponding to abnormal
   learning, memory, and conditioning phenotype (Monarch 2063). Comparing
   R55-induced changes to LTP time-course datasets [[80]5], we found that
   R55-induced expression changes were strongly aligned with late-phase
   (2 h) LTP-associated transcriptional signatures (Fig. [81]4A). For most
   of these genes the expression changes induced by LTP is opposed to the
   AD pathological expression changes in the MODEL-AD dataset.
   Interestingly, the R55 treatment mostly resulted in gene expression
   changes in the same direction as late LTP, counteracting the 5xFAD
   changes (Fig. [82]4B). This suggests that R55 treatment counteracts the
   synaptic and LTP dysfunction observed in 5xFAD mice.

Fig. 4.

   [83]Fig. 4
   [84]Open in a new tab

   Retromer stabilization restores synaptic gene expression in 5xFAD mice.
   A Correlation of gene expression changes in genes related to abnormal
   learning, memory and conditioning in R55-treated 5xFAD mice (vs.
   vehicle) with those observed following chemically induced long-term
   potentiation (cLTP) at 5, 60, and 120 min, as reported by Bliim et al.
   [[85]5]. B Expression changes of the same genes in our dataset, the
   Bliim et al. dataset, and the MODEL-AD dataset at 4 and 8 months,
   providing a temporal context for synaptic gene regulation in AD
   progression

   To summarize and visualize the changes caused by R55 and R33 treatment,
   we mapped differential expression onto a pathway diagram of endosomal
   processing in AD (Fig. [86]5). Interestingly, core retromer proteins
   (Vps35, Vps26a, Vps29) remained unchanged following R55 treatment and
   during disease progression (MODEL-AD), but their known cargos and
   associated proteins showed widespread alterations. This implies
   functional changes to the retromer that are not through expression
   changes, that has extensive effects on the endosomal system and many AD
   related cellular pathways. Notably, both R33 and R55 treatment resulted
   in consistent and significant changes in the entire Vps13
   domain-containing family, which are critical for endosomal function.

Fig. 5.

   [87]Fig. 5
   [88]Open in a new tab

   Summary of alterations in the retromer complex and related proteins
   relevant for Alzheimer's disease. This manually curated diagram
   summarizes transcriptional changes in retromer-associated genes and
   related proteins implicated in AD following treatment with R33 (left
   area in the box for each gene) and R55 (right area). Gene expression
   changes are represented using a color scale, where green indicates
   downregulation and red indicates upregulation, with color intensity
   proportional to the magnitude of change. Genes with minimal or
   statistically insignificant alterations are displayed in near-white for
   completeness. Unlike Figs. [89]3, [90]4, and [91]8, which show log₂
   fold changes relative to vehicle controls, this figure presents mean
   Z-normalized values for each treatment in the injected hemisphere,
   similar to the heatmap in Fig. [92]2C

Retromer stabilization alters subcellular protein localization

   Given the unexpected alterations in Vps13 gene family expression, we
   investigated whether retromer stabilization impacted subcellular
   protein distribution. As expected, VPS35 protein levels did not differ
   significantly between 5xFAD mice and WT litter mates (supplementary
   Fig. [93]2). Immunohistochemistry of VPS35, VPS13b, and RAB7 in the R55
   microinjected cortex revealed a significant reduction in VPS13 protein
   levels, an increase in RAB7 and VPS35 protein levels remained unchanged
   (Fig. [94]6), all consistent with the gene expression results. In
   vehicle-treated 5xFAD mice, VPS35-positive vesicles were irregularly
   distributed, often near the nucleus. However, in R55-treated mice,
   VPS35-positive vesicles formed dense clusters and were significantly
   shifted away from the nucleus (Fig. [95]6), indicative of enhanced or
   restored endosomal maturation, potentially driven by normalized
   expression of VPS13 and RAB7. To determine whether the increase in Rab7
   expression was accompanied by changes in lysosomal biogenesis, we
   examined the transcriptional expression of genes linked to the
   TFEB/TFE3 axis and lysosomal degradation [[96]32]. No significant
   changes were observed in transcriptional markers of this pathway, such
   as ATG7, ATG14, WIPI1, ULK1, ULK2, or SQSTM1 [[97]44], supporting the
   interpretation that lysosomal biogenesis was not enhanced after R55
   treatment. To further examine the effect on retromer function we
   analysed the size and subcellular distribution of Sorl1 labelled
   particles after vehicle and R55 treatment (Fig. [98]7A) and plotted the
   data in population density maps (Fig. [99]7B). In 5XFAD vehicle-treated
   brains, Sorl1 particle size is distributed in a wide range between 0.1
   and 0.3 μm^2 with peaks at 0.15 and 0.36 μm^2. When measuring the
   distance from the nuclear envelope, the main peak is at 5.2 μm, there
   is a second maximum at 1.7 μm consisting of larger particle sizes.
   Interestingly, in R55 treated brains this second peak is absent.
   Furthermore, in the vehicle-treated brains almost half of the particles
   (5th contour line on the overlap) are less than 1 μm from the nuclear
   envelope while in the R55-treated brains, only 20% of the particles
   (3rd contour line on the overlap) remain at this close position. In
   addition, the particle sizes are larger in the vehicle-treated brains.
   In summary, R55 treatment results in a normalization of Sorl1 particle
   sizes and distributions.

Fig. 6.

   [100]Fig. 6
   [101]Open in a new tab

   Immunohistochemical analysis of VPS35, VPS13, and RAB7 following R55
   treatment in 5xFAD mice. A Representative images of VPS35
   immunolabeling (red) in vehicle- and R55-treated mice (top). Nuclei
   labeled by DAPI. Mid panel shows the corresponding 3D reconstructions
   of the VPS35 immunolabeling. Quantification of the mean fluorescent
   intensity (arbitrary units) of VPS35-positive particles (bottom) shows
   no change in the amount of VPS35, but that R55 treatment reduce the
   distance to the nearest particle (increased VPS35 clustering) and the
   spatial distribution away from the nucleus. B Representative images of
   VPS13 (green) and RAB7 (red) immunolabeling in vehicle- and R55-treated
   mice (top). Quantification of VPS13 and RAB7 mean fluorescent intensity
   (arbitrary units) within segmented region of interest (bottom) confirms
   R55-induced VPS13 reduction and RAB7 upregulation. Note that background
   staining, most evident in the R55 image was not included in the
   segmentation which was done around each nucleus. C Expression changes
   of genes linked to the TFEB/TFE3 axis and lysosomal degradation in our
   dataset (green). No significant changes were observed in ATG7, ATG14,
   WIPI1, ULK1, ULK2, or SQSTM1 (red line shows p = 0.05), supporting the
   interpretation that lysosomal biogenesis was not enhanced after R55
   treatment. MODEL-AD dataset at 4 and 8 months included for comparison

Fig. 7.

   [102]Fig. 7
   [103]Open in a new tab

   Retromer stabilization alters the size and cellular distribution of
   SORL1 labeled particles in a manner consistent with more efficient
   retromer function. A Representative Sorl-1-stained (green) images of
   R55-treated and vehicle-treated 5xFAD mice. B Density map of Sorl-1
   particle size (horizontal axis) versus distance to the nuclear envelope
   (vertical axis). Increasingly warmer colors indicate increasing numbers
   of particles with each axis feature. In the vehicle-treated 5xFAD mice
   there is a significant number of larger particles that have travelled
   less (right yellow peak and shallow slope below), which is absent in
   the R55-treated mice. The curves represent the increase in population
   density from 0 to 90% in 10% steps, plus an additional step at 95%. The
   X's showing the local maxima are to aid interpretation of the next
   panel. C Overlay of the two 2D density maps and lateral projections on
   distance (vertical axis) and size (horizontal axis), color coded for
   vehicle and R55 treated 5xFAD mice. The Xs on the 2D maps are shown as
   dashed lines on the lateral 1D densities

Retromer stabilization modulates microglial activation

   As VPS35 plays a critical role in microglial activation [[104]33], we
   investigated its impact on neuroglial interactions. Gene expression
   analysis after retromer stabilization revealed activation of microglial
   phenotype-associated genes, changes in astroglial receptors,
   inactivation of glutamatergic synapses and activation of some genes
   downstream of Creb in neurons. These changes caused by R55 and R33
   treatment on neuroglial interactions with a focus on the glutamatergic
   synapse are summarized and visualized in an annotated pathway diagram
   (Fig. [105]8).

Fig. 8.

   [106]Fig. 8
   [107]Open in a new tab

   Retromer stabilization alters glutamatergic synapses, astrocyte
   function, and microglial activation. This diagram, adapted from WP5083,
   illustrates transcriptional changes in glutamatergic synapses,
   astrocytes, and microglia following treatment with R33 (left area in
   the box for each gene) and R55 (right area). Gene expression changes
   are represented using a color scale, where green indicates
   downregulation and red indicates upregulation, with color intensity
   proportional to the magnitude of change. Genes with minimal or
   statistically insignificant alterations are displayed in near-white for
   completeness. As in Fig. [108]5, this figure presents mean Z-normalized
   values for each treatment in the injected hemisphere, rather than log₂
   fold changes relative to vehicle controls, as shown in Figs. [109]3,
   [110]4, and 8

   We further investigated the effect of retromer stabilization on
   microglia using IBA1 immunohistochemistry, this showed that microglia
   in R55-treated 5xFAD mice exhibited reduced ramification
   (Fig. [111]9A), suggesting further activation of microglia. To
   investigate these effects we compared the R55-induced differential
   expression changes with those reported by Keren-Shaul et al. [[112]19]
   on FACS-sorted (CD45 +) immune cells from 5xFAD brains, characterizing
   the genes responsible for the homeostatic to DAM, and DAM I to DAM II
   phenotype. We intersect their microglia gene set with our R55 vs.
   vehicle data, and plotted the changes we observed against their
   changes, as well as the changes in MODEL-AD (Fig. [113]8B). The left
   panel compares our results in the genes from the Keren-Shaul dataset to
   MODEL-Ad at 4 months (Log2FC.R55 vs. Log2FC.m4). Genes that are
   compensated by R55 occupy the upper left and lower right quadrants and
   are coloured red. Genes with changes of the same sign in both models
   are coloured green. In the middle panel, R55 changes are plotted
   against the Keren-Shaul's changes from homeostatic microglia to DAM
   (log2FC.R55 vs logFC.DAMfromHOM). The colour coding of each gene is
   retained from the first diagram. This shows that a large majority of
   the originally"red"genes now change with the same sign, thus being
   consistent with microglial activation after R55 treatment. Next, we did
   the same against the changes from DAM I to DAM II2 (log2FC.R55 vs
   log2FC.DAM2fromDAM1), with similar results.

Fig. 9.

   [114]Fig. 9
   [115]Open in a new tab

   Retromer stabilization modulates microglial activation states in 5xFAD
   mice. A Representative IBA-1-stained (green) images of microglia in
   R55-treated and vehicle-treated 5xFAD mice. Nuclei labeled by DAPI.
   Microglia appear less ramified and complex following R55 treatment. B
   Comparison of the changes on drug treatment versus microglial
   activation. To localize the genes putatively responsible for the
   changes quantitatively observed in (A), we compared our changes in
   R55/vehicle-treated 5xFAD mice (log2FC R55) with differential
   expression on the microglial activation model described in
   [[116]https://doi.org/10.1016/j.cell.2017.05.018], which includes
   differential expression from homeostatic microglia to DAM (logFC2
   DAMfromHOM), differential expression from DAM I to DAM II (logFC2
   DAM2fromDAM1). We also included the changes in MODEL-AD at 4 months
   (logFC2 m4). For this comparison, we only used genes with adjusted
   p < 0.05 in our treatment that were also present in the microglia
   study, so mostly microglia genes. Genes whose changes are of the same
   sign, both increased in R55 over vehicle and 5xFAD over WT, or both
   reduced, are shown in green. C Statistical modelling of (B). To
   statistically analyze the relevance of microglial activation in the
   changes observed after R55, we build a series of nested general linear
   models whose AIC (lower the better) relates to the goodness of fit. The
   model that includes both MODEL-AD and microglial activation has a
   better AIC and is significantly better than the others (likelihood
   ratio test vs. MODEL-AD alone: p = 0.003). We then calculated the
   contributions of each factor, including interactions, which can be
   observed in the right panel. The first three are significant. D Changes
   in microglial DAM-related gene expression following R55 treatment,
   contextualized against MODEL-AD datasets at 4 and 8 months

   Finally, to obtain a statistical quantification, we build a series of
   generalized linear models of increasing complexity that express the
   changes we observed in R55/veh (log2FC.R55) as a function of the data
   in plot 8B separately (logFC.m4, log2FC.DAMfromHOM,
   log2FC.DAM2fromDAM1) or in combination. We calculated the Akaike
   information criterion (AIC) for each model and found that the model
   that included all factors had the best AIC (8C) and was statistically
   different from the others on the likelihood ratio test. We then
   estimated the statistical contribution of each parameter to the model
   and the combinations between them, and found that logFC.m4,
   log2FC.DAMfromHOM, log2FC.DAM2fromDAM1 contribute significant
   separately, and that log2FC.DAMfromHOM actually has a greater effect
   than log2FC.m4. The combination of microglial activation phenotypes
   (log2FC.DAMfromHOM * log2FC.DAM2fromDAM1) almost reaches statistical
   significance. Taking this all together, shows that the DAM from HOM and
   DAMII from DAM I changes explain a significant part of the changes we
   observe in microglial genes after R55 treatment (Figs. [117]9C, D).

Discussion

Retromer stabilization as a potential strategy to reverse AD-related
molecular alterations

   Disruptions in synaptic transmission and plasticity are early hallmarks
   of AD. Endosomal trafficking, mediated by the retromer complex, plays a
   pivotal role in synaptic function by regulating intracellular protein
   transport. Our study demonstrates that stabilizing the retromer complex
   reverses key molecular alterations in AD, shedding light on downstream
   mechanisms and signaling pathways affected by retromer dysfunction. Our
   results supports previous reports that dysfunctional retromer linked to
   neurodegenerative diseases is not primarily caused by loss of
   expression of Vps35 but by decreased endosomal association causing loss
   of function that can be mitigated by retromer stabilizing compounds
   such as R55 and R33 [[118]10]. Additionally, our analysis confirms that
   5xFAD mice exhibit extensive gene expression changes in the retromer
   interactome and its cargo, resembling patterns seen in human AD cohorts
   — a degree of alteration not observed in other transgenic AD models.

Retromer stabilization restores synaptic plasticity and LTP-associated
pathways

   A strong body of evidence links synaptic dysfunction to early AD
   progression, with Aβ oligomers interfering and inducing changes similar
   to those involved in developmental synaptic pruning and with
   plasticity-related processes such as LTP and LTD [[119]23]. Under
   normal conditions, these interactions are transient, but in AD, Aβ
   oligomers induce persistent synaptic alterations, contributing to
   disease progression. For example, Aβ oligomers increase cholesterol
   concentration at synapses, promoting Aβ/PrP complexes that disrupt
   NMDAR signaling via GRIN2a/GRIN2b S-nitrosylation [[120]3]. In line
   with these events, we observed compensatory increases in Grin2a,
   Grin2b, iNos, and VLDL/ApoE-related genes such as Lrp1, Lrp1b, and Fbn1
   in 5xFAD mice (MODEL-AD datasets). Notably, retromer stabilization by
   R55 counteracted these transcriptional upregulations, suggesting
   enhanced receptor recycling and trafficking reduced the need for
   compensatory gene expression changes.

   The retromer complex is crucial for synaptic AMPA receptor trafficking.
   Loss of VPS35 impairs Gria1 (AMPA) receptor insertion and disrupts LTP,
   while restoring VPS35 reverses these deficits [[121]39]. In our study,
   R55 treatment normalized Gria1 and Grip1 expression, the latter being
   essential for AMPA receptor endosomal sorting and membrane targeting.
   This is particularly significant given recent findings implicating the
   semaphorin family (Plexin a4, Sema5a) and netrin receptors (Unc5c, Dcc)
   in AMPA receptor targeting during LTP-associated plasticity [[122]17].
   In 5xFAD mice, these genes were dysregulated, but their expression was
   restored by retromer stabilization, further supporting a role for
   retromer function in synaptic integrity.

Retromer stabilization modulates calcium handling in AD pathology

   Defective intracellular calcium homeostasis, particularly endoplasmic
   reticulum (ER) calcium leakage, is an early event in AD-linked synaptic
   dysfunction. The neuronal ryanodine receptor (Ryr2) is overactivated in
   AD, leading to enhanced receptor-operated calcium entry and decreased
   buffering capacity [[123]18, [124]21]. Furthermore, RyRs interact
   functionally with L-type calcium channels (Cacna1c, Cacna1d, Cacna1e),
   particularly in CA1 pyramidal neurons, where calcium-induced calcium
   release contributes to synaptic plasticity [[125]35].

   In 5xFAD mice, we observed elevated Ryr2 and Cacna1c/Cacna1d/Cacna1e
   expression, consistent with previous findings. R55 treatment
   effectively reversed these alterations, suggesting restored calcium
   signalling homeostasis. Of Ryr2-associated regulators [[126]38], only
   Jph4 showed a weak response to R55, indicating selective rather than
   global modulation of calcium-handling genes.

Effects of retromer stabilization on Aβ processing and endosomal function

   Given that the retromer complex regulates Aβ processing, we examined
   key AD-related cargos. R55 treatment increased γ-secretase components
   while reducing Sorl1 and Col25a1 expression. Col25a1 encodes a
   collagen-like amyloid plaque component, known for forming
   protease-resistant aggregates. These changes suggest that recovered
   retromer function may reduce excessive amyloidogenic processing caused
   by extended cargo residency at the Golgi by normalizing Golgi-retromer
   dynamics.

   Unexpectedly, R55 treatment significantly reduced Vps13 family gene
   expression, which was otherwise elevated in 5xFAD mice. The Vps13
   complex regulates endosomal recycling towards the Golgi, mitochondria,
   and autophagosomes by promoting organelle bilayer bridging, and has
   been implicated in mitochondrial dysfunction in neurodegenerative
   diseases, including atypical PARK23 [[127]40]. Genetic variations in
   VPS13 genes are emerging risk factors for sporadic, late-onset AD, with
   VPS13 variants exceeding even ApoE in risk contribution [[128]27].
   Immunohistochemistry confirmed that Vps13 protein levels decreased
   following R55 treatment, while RAB7 expression increased, this suggests
   restored endosomal maturation. Previous studies have shown a stagnation
   of large endosomes both in AD and in 5xFAD [[129]9, [130]24] and
   impaired recycling pathway, in 5xFAD in mice of almost the same age
   [[131]24]. We used SORL1 expression as a marker of endosomal
   trafficking. SORL1 expression has been shown to have a reciprocal role
   prototypical cargos such as transferrin to in cell surface recycling
   [[132]30]. Sorl1 distribution that has been reported to be altered both
   in APP overexpressing AD models such as 5xFAD and in Sorl1 mutations
   that have been linked to AD pathology [[133]29]. The changes to Vps13
   and RAB7, taken together with the increased proximity of both VPS35 and
   SORL1 to the plasma membrane, compensation in synaptic-related
   pathways, and the absence of changes in lysosomal
   degradation-associated genes, our data support improved sorting and
   membrane trafficking that may favor recycling over degradation.

Retromer stabilization modulates neuroinflammation and microglial
polarization

   In the neuroinflammatory domain, R55 treatment increased both
   pro-inflammatory mediators (Il1, Nfkb2) and pro-trophic glial markers
   (Il4r, Il13ra1), as well as downstream anti-inflammatory transcription
   factors (Stat6, Stat3). This coordinated shift in crosstalk between
   microglial polarization and astrocytic responses mirrors findings
   reported in other AD studies [[134]45]. Similarly, Vps35 KO impaired
   microglial activation in response to Aβ [[135]33].

   IBA1 staining revealed that R55-treated 5xFAD microglia exhibited
   reduced ramification, indicating an altered activation state. Further
   gene expression comparisons to MODEL-AD and DAM datasets showed
   significant shifts in microglial phenotypes [[136]8, [137]15]. Given
   that phagocytic receptors and cytokine receptors utilize
   retromer-mediated recycling in glia, these findings provide a
   mechanistic link between retromer function and neuroinflammation
   although further studies are needed to assess if the identified changes
   have long term positive or negative effects.

Limitations of the study

   This study focuses on dysregulation of the transcriptome in early AD
   and the acute molecular effects of retromer stabilizing drugs. Although
   our results suggests that retromer dysfunction can contribute to many
   AD related alterations and that the retromer stabilizing drugs have the
   potential to counteract these effects, further studies are needed to
   confirm these findings. Comparing the changes we identify to downstream
   effects on the proteome would be interesting, but large-scale
   proteomics are still challenging and currently there are no database
   for changes to the proteome similar to the MODEL-AD. As the focus of
   this study was on acute changes, we can only speculate on the long-term
   effects on AD-related phenotypes such as memory, cognition, and
   synaptic plasticity. Further studies assessing pharmacokinetics and
   pharmacodynamics of this type of drugs are also needed.

Implications and future directions

   While our study provides acute mechanistic insights, extrapolating
   these findings to chronic AD progression remains challenging. However,
   our results strongly support retromer dysfunction as a key contributor
   to synaptic pathology in AD, raising the possibility that retromer
   stabilization could mitigate disease progression.

   One intriguing question is why VPS35 mutations (e.g., D620N) cause
   Parkinson’s disease (PARK17) rather than AD. The D620N mutation
   disrupts VPS35 binding to WASH complex proteins, affecting actin
   nucleation, while AD-linked VPS35 mutations (e.g., L625P) impair
   VPS35-VPS26-VPS29 assembly [[138]25, [139]34]. Notably, R55 enhances
   VPS35 assembly without affecting WASH binding [[140]4]. However,
   R55-treated 5xFAD mice exhibited changes in WASH complex components,
   sorting nexins (Snx1, Snx2, Snx3), RAB7, and Vps13, suggesting a
   broader role for retromer stabilization in intracellular trafficking
   beyond cargo selection.

   Our findings suggest that retromer stabilization influences not only
   endosomal cargo selection but also endosomal anchoring to other
   organelles. The observed redistribution of VPS35-positive vesicles away
   from the nucleus suggests that endosomes are repositioned towards the
   plasma membrane, likely improving membrane protein recycling, the
   decrease in Vps13 and increase in RAB7 further support this hypothesis.

Conclusion

   We have identified a range of retromer-regulated cargos, including
   membrane receptors and synaptic proteins, that are dysregulated in
   early AD and restored by retromer stabilization. Our data suggest that
   retromer dysfunction contributes to synaptic and neuroinflammatory
   alterations in AD, and that stabilizing the retromer complex enhances
   endosomal recycling, synaptic receptor trafficking, and microglial
   activation states. As 5xFAD mice lack genetic alterations in VPS35 or
   other retromer components, our findings may be generalizable to
   sporadic AD cases, supporting the exploration of retromer-directed
   therapies as a potential disease-modifying strategy.

Supplementary Information

   [141]Supplementary Materials^ (2.5MB, pdf)

Acknowledgements