Single-Nucleus Transcriptomics Analysis Identifies Sex-Dichotomous Pathological Axes in Alzheimer’s Disease: Female Homeostatic Failure versus Male Neuroimmune Activation

Single-Nucleus Transcriptomics Analysis Identifies Sex-Dichotomous Pathological Axes in Alzheimer’s Disease: Female Homeostatic Failure versus Male Neuroimmune Activation | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Single-Nucleus Transcriptomics Analysis Identifies Sex-Dichotomous Pathological Axes in Alzheimer’s Disease: Female Homeostatic Failure versus Male Neuroimmune Activation Ziyi Zhou, Chenxi Jiang, Zhenzhen Chen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9504119/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Alzheimer’s disease (AD) disproportionately affects females in both prevalence and clinical progression, yet the underlying sex-dependent cellular programs remain poorly understood. This study aims to delineate the sex-specific pathological landscape of AD at a single-nucleus resolution to identify the biological drivers of these disparities. We analyzed a large-scale integrated snRNA-seq dataset from the human prefrontal cortex, encompassing 179,690 nuclei harmonized across three foundational AD cohorts. Results Our analysis revealed a sex-dichotomous pathological landscape. In females, AD is characterized by a systemic erosion of homeostatic programs: microglial oxidative phosphorylation drops sharply, astrocytic APOE expression is significantly downregulated, and the presynaptic compartment is severely depleted. In contrast, males exhibit a distinct pathological axis dominated by sustained neuroimmune activation while metabolic and synaptic capacities remain relatively preserved. Furthermore, we identified a functional collapse of the estrogen signaling axis in females; despite a paradoxical upregulation of ESR1 , its downstream neuroprotective targets remain inactive. The female-specific downregulation of astrocytic APOE potentially explains the amplified genetic risk associated with the APOE ε4 allele in women. Conclusions Our findings demonstrate that biological sex is a primary determinant of cellular vulnerability in AD. The identified transcriptional uncoupling of estrogen signaling and sex-specific metabolic-immune failure underscore the necessity of sex-stratified approaches in AD precision medicine and the design of targeted interventions. Alzheimer’s disease snRNA-seq Sex differences Homeostatic failure Neuroimmune activation Microglial metabolism Astrocytic APOE Estrogen signaling decoupling Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Introduction Alzheimer’s disease (AD) is the most prevalent neurodegenerative disorder worldwide and constitutes a rapidly escalating public health burden in aging societies [ 1 ]. Therapeutic development has focused intensively on canonical pathological hallmarks, including beta-amyloid plaques and intracellular neurofibrillary tangles, yet these methods have gained limited clinical success. The majority of late-stage trials that target these pathways have failed to prove meaningful efficacy [ 2 ]. One factor underlying this persistent translational impasse is the profound clinical and biological heterogeneity of AD. It has thus become urgent to identify the biological modifiers shaping this heterogeneity and to integrate them into precision medicine frameworks. Among such modifiers, biological sex has gained substantial recognition as a determinant of AD risk, progression, and molecular pathology [ 3 ]. Epidemiological and clinical studies demonstrate pronounced sex differences in AD prevalence and disease trajectory. Women account for approximately two-thirds of sporadic AD cases, indicating a markedly higher lifetime risk compared to men [ 1 , 4 ]. Longitudinal studies demonstrate that women diagnosed with AD experience a more rapid rate of cognitive decline than their male counterparts [ 5 ]. Genetic susceptibility also exhibits sex-dependent effects. Meta-analyses have shown that female carriers of the apolipoprotein E ε4 (APOE ε4) allele, particularly between the ages of 65 and 75, face a significantly higher risk of developing AD than male carriers of the same genotype [ 6 ]. Multiple biological mechanisms likely contribute to this disparity. At the genetic level, APOE ε4 exerts a stronger risk-amplifying effect in women within specific age windows, while the potentially protective APOE ε2 allele may confer greater benefit in women than in men. Endocrine factors also play a pivotal role. Women experience a relatively abrupt decline in circulating estrogen levels during the menopausal transition, in contrast to the more gradual age-related decline in androgens in men. Clinical investigations support the “critical window hypothesis” for hormone replacement therapy (HRT), suggesting that neuroprotective effects may only be achievable when therapy is initiated during the perimenopausal period. However, large randomized controlled trials, including the Women’s Health Initiative Memory Study (WHIMS), demonstrated that combined estrogen-plus-progestin therapy increases the risk of dementia and mild cognitive impairment (MCI) in postmenopausal women aged 65 years and older [ 7 ]. Estrogen-only therapy likewise failed to reduce dementia risk and may increase the combined incidence of dementia and MCI in this population [ 8 ], leading to recommendations against HRT for dementia prevention. Neuroimmune mechanisms, particularly those involving microglia, further contribute to sex-specific vulnerability in AD. Experimental studies indicate that microglia from aged female models exhibit enhanced basal phagocytic capacity for neuronal debris but display impaired dynamic regulation of this function in response to inflammatory stimuli [ 9 ]. In human postmortem brain tissue, estrogen receptor alpha (ESR1) expression positively correlates with the microglial phagocytosis-associated receptor TREM2 in samples from postmenopausal women, suggesting a potential link between estrogen signaling and microglial function in the aging female brain [ 10 ]. Consistent with these molecular findings, neuroimaging studies reveal that women may exhibit greater hippocampal atrophy at Aβ burden levels comparable to those at which men remain relatively stable [ 11 ]. Furthermore, women exhibit stronger associations between global Aβ burden, regional tau pathology, and cognitive decline than men at similar levels of amyloid deposition [ 12 ]. Despite these advances, major gaps remain in understanding the molecular and cellular mechanisms driving sex differences in AD. Current research examining interactions among estrogen signaling, APOE genotype, and neuroinflammatory pathways remains fragmented, and their combined or cell-type-specific effects are poorly resolved [ 3 , 9 , 10 ]. Critically, there is a lack of systematic analysis characterizing sex-specific transcriptional programs and functional states across diverse brain cell types (e.g., neurons, astrocytes, microglia) at single-cell resolution in human AD brains [ 3 , 10 ]. Furthermore, commonly used animal models inadequately capture the prolonged human perimenopausal transition and the complex temporal dynamics of brain aging, limiting their translational relevance for studying sex-dependent mechanisms [ 3 , 9 ]. To address these limitations, systematic bioinformatic analyses leveraging high-quality human single-cell datasets stratified by sex are essential for elucidating the cellular and molecular basis of sexual dimorphism in AD. In this study, we analyzed an integrated human single-nucleus RNA sequencing (snRNA-seq) dataset published by Li et al. (2025) in Cell [ 13 ], which harmonizes data from three foundational AD cohorts Mathys et al. (2019) [ 14 ], Zhou et al. (2020) [ 15 ], and Lau et al. (2020) [ 16 ] using rigorous cross-platform normalization and batch correction. This dataset provides a robust framework for cell-type resolved investigation of AD-associated transcriptional changes. Unlike the original study, we explicitly model AD transcriptional programs through a sex-stratified and sex-by-diagnosis framework across major brain cell types, integrating differential expression, pathway coordination, functional module scoring, and network-level signaling inference to delineate sex-dependent cellular states. Results Integrated cohort validation and confounder control To systematically investigate sex-stratified cellular pathology in AD, we leveraged the pre-processed, harmonized single-nucleus RNA sequencing (snRNA-seq) dataset of the human prefrontal cortex published by Li et al. (2025) [13], which integrates three foundational AD cohorts [14-16]. We curated the dataset by retrieving and appending complete age metadata manually from the original studies, achieving a final cohort of 75 samples (37 AD, 38 control [CTL]) with 100% completeness for key clinical variables. CERAD score, demonstrating that age does not confound pathological severity in this dataset. Systematic cohort characterization validated robust data, which were consistent with expected neuropathological features (Fig. 1). As anticipated, AD cases exhibited significantly higher Braak stages (Fig. 1b, e) and lower CERAD scores (Fig. 1c, f) compared to controls, validating diagnostic annotation. Within each sex, age distribution did not differ significantly between AD and CTL groups (Fig. 1g-i), and APOE ε4 allele frequency followed established disease-risk patterns (Fig. 1d). No significant sex-by-age interaction was detected within either diagnostic group, supporting the absence of major demographic confounding. Together, these results establish a well-balanced cohort that provides a solid foundation for downstream sex-stratified, cell-type-specific analyses. Cellular atlas reveals sex-dimorphic shifts in cell-type proportions in AD We analyzed 179,690 high-quality nuclei from the integrated dataset and resolved six major neural cell populations (Fig. 2a). For clarity, we refer to these as excitatory neurons (ExN), inhibitory neurons (InN), astrocytes (Ast), microglia (Mic), oligodendrocytes (Oli), and oligodendrocyte precursor cells (OPC); corresponding labels in figures are ex.neu, in.neu, ast, mic, oli, and opc, respectively. Uniform manifold approximation and projection (UMAP) visualization demonstrated effective integration across source cohorts, sexes (Fig. 2b), and diagnostic groups (Fig. 2c), with no gross spatial segregation observed in sex-resolved AD or control subsets (Fig. 2d, e). Cell-type annotations were further validated by canonical marker gene expression (Fig. 2h). Comparative analysis of cell-type proportions uncovered subtle yet statistically significant alterations in AD that diverged by sex (Fig. 2f-g, i). In both sexes, AD was associated with an increased proportion of astrocytes and a reduction in inhibitory neurons. Strikingly, excitatory neurons and oligodendrocytes exhibited opposing trajectories: female AD samples showed an increase in the proportion of excitatory neurons accompanied by a decrease in oligodendrocytes, whereas male AD samples displayed the inverse pattern. Though such changes may reflect selective cellular vulnerability or survival, transcriptional state changes influencing cell-type assignment cannot be fully excluded and are addressed in subsequent functional analyses. Collectively, these findings suggest that AD engages distinct cellular remodeling programs in males and females. Transcriptomic profiling uncovers sex-polarized functional dysregulation To sensitively capture molecular alterations, gene filtering thresholds were recalibrated to retain genes expressed in at least 3 cells, maximizing detection power for differential expression analysis (see Methods). Cell-type-stratified analysis between AD and CTL samples was performed separately for each sex. Across all cell types, differential expression analysis revealed widespread transcriptional alterations exhibiting minimal cross-sex overlap among differentially expressed genes (DEGs) (Fig. 3a, b). Venn diagrams and volcano plots for each cell type quantified this divergence, demonstrating that the majority of DEGs were sex-specific (Fig. 3c-h; Fig. 4a-f). Importantly, this sex specificity was not driven by differences in cell number or sequencing depth, supporting a biological rather than technical origin of the observed polarization. Gene Ontology (GO) enrichment analysis of female-specific DEGs revealed a coherent pattern of disruption in pathways critical for intracellular homeostasis and neuronal functions (Fig. 4g, h, i, k). In females, microglial DEGs were overwhelmingly enriched for mitochondrial electron transport and oxidative phosphorylation. Oligodendrocyte DEGs were associated with RNA splicing and microtubule polymerization, while excitatory neuron DEGs were enriched for synaptic vesicle cycling, neurotransmitter secretion, and axonogenesis. Together, these signatures point to a concerted impairment of bioenergetic, post-transcriptional regulatory, and synaptic machinery in the female AD brain. In contrast, GO analysis of male-specific DEGs highlighted a distinct pathological axis centered on immune activation and inflammatory signaling (Fig. 4j, l). Male microglial and astrocyte DEGs were significantly enriched for pathways related to T cell differentiation, lymphocyte activation, and cytokine production, indicating a pronounced neuroinflammatory response. These sex-divergent functional programs were further supported by complementary KEGG pathway analysis (Supplementary Fig. S1). Focused examination of high-priority genes implicated in AD genetics and the identified functional signatures, including immune regulators (e.g., TREM2 , TYROBP ), the major genetic risk factor APOE , synaptic genes (e.g., SNAP25 , SYT1 ), and the estrogen receptor ESR1 . Their cell-type-resolved expression profiles confirmed the profound sex-specificity of AD-associated transcriptional alterations (Fig. 5a, b). In females, AD was characterized by a pronounced downregulation of synaptic genes, particularly SNAP25 and SYT1 , most evident in oligodendrocytes and oligodendrocyte precursor cells. APOE expression decreased in astrocytes but increased in microglia, while ESR1 expression was elevated across multiple cell types. In males, changes in these genes were comparatively attenuated. Detailed expression statistics are provided in Supplementary Table 3. Estrogen signaling pathway is dysregulated in female AD Given the pronounced female-specific transcriptional alterations, we investigated estrogen signaling using a curated gene set encompassing nuclear receptors (ESR1, ESR2, PGR), primary transcriptional targets (GREB1), and downstream effectors involved in neuroprotection (BDNF, SIRT1, BCL2), cell proliferation (CCND1, MYC), and neuronal function (NRCAM, VEGFA) (Fig. 6e). Pathway coordination was first assessed via pairwise Spearman correlations among estrogen pathway genes in female brain cells. Consistent with prior single-cell studies, absolute correlation coefficients were modest; however, pathway-level coordination was statistically robust in both AD and control groups (all adjusted p < 0.05; Fig. 6a). Differential correlation analysis revealed substantial rewiring of ESR1-centered interactions in AD (Fig. 6b, c). Correlations between ESR1 and several pathway components were attenuated (e.g., ESR2, SIRT1), whereas the association with the cell cycle regulator CCND1 was strengthened. This occurred alongside a generalized increase in ESR1 expression across female AD cell types (Fig. 5), suggesting compensatory upregulation amid impaired pathway coordination. Functional validation using Gene Set Variation Analysis (GSVA) demonstrated broad dampening of early and late estrogen response programs in female AD brains, most prominently in astrocytes and OPCs (Fig. 6d). In contrast, male AD samples exhibited minimal and inconsistent changes in these pathways (Fig. 6f). Together, these findings delineate a distinct impairment of estrogen signaling networks in female AD, characterized by discordant gene–gene associations and lowered functional output despite elevated ESR1 expression. Astrocytic APOE reduction is linked to metabolic-immune dysregulation in female AD Given the central role of APOE as the strongest genetic risk factor for AD, we further dissected its sex- and cell-type-specific contributions. APOE expression was predominantly localized to astrocytes, with lower expression in microglia and minimal expression in neurons (Fig. 7a). A significant reduction in APOE expression was observed specifically in female AD astrocytes, accompanied by increased expression in female microglia. Male samples showed similar but notably attenuated trends. Genome-wide correlation analysis in female astrocytes revealed extensive co-dysregulation associated with APOE expression (Fig. 7b). Gene set enrichment analysis (GSEA) revealed significant enrichment of pathways related to energy metabolism, lipid handling, oxidative stress, complement activation, and inflammatory signaling (all FDR < 0.001; Fig. 7c, d). While causality cannot be inferred from correlation-based analyses, the strong coupling of astrocytic APOE reduction with metabolic and immune pathways suggests a central role for astrocytic dysfunction in female AD vulnerability. The concomitant increase of APOE in microglia may reflect compensatory redistribution of functional burden across glial compartments. Microglial functional states exhibit sex-dichotomous dysregulation To characterize microglial functional alterations, module scores were calculated for metabolic activity, clearance capacity, and immune activation (Fig. 8a). Female AD microglia displayed marked metabolic impairment compared to controls (p < 0.0001), accompanied by a modest but significant reduction in clearance capacity (p < 0.01). In contrast, male AD microglia showed preserved metabolic function but a pronounced increase in immune activation relative to male controls (p < 0.0001). A weak but significant positive correlation between metabolic and clearance scores was detected specifically in female AD microglia (Pearson’s r = 0.10, p = 5.03 × 10⁻⁵; Fig. 7c). Gene-level analysis supported these functional patterns: male AD microglia exhibited modest downregulation of the homeostatic marker P2RY12 and upregulation of TREM2, whereas female AD microglia showed elevated APOE expression (Fig. 8b). Together, these results reveal a sex-dependent inversion of microglial functional states in AD, with female microglia characterized by metabolic insufficiency and males by heightened inflammatory activation. Excitatory neurons exhibit sex-divergent synaptic gene dysregulation Motivated by synaptic pathway enrichment in female excitatory neuron DEGs, we analyzed a focused panel of synaptic genes encompassing presynaptic vesicle release, postsynaptic transduction, and synaptic adhesion. In pooled AD samples, presynaptic genes (SYP, SNAP25, RAB3A, VAMP2) were predominantly downregulated, whereas several postsynaptic genes ( GRIN1, DLG2 ) were upregulated (Fig. 9a, b). Sex-stratified analysis revealed that this presynaptic–postsynaptic imbalance was substantially more pronounced in females. Female AD samples exhibited severe downregulation of vesicle release machinery, particularly VAMP2 and RAB3A, whereas alterations in male AD samples were comparatively mild (Fig. 9d, f). Expression distributions further highlighted increased heterogeneity and pronounced presynaptic loss specifically in female AD excitatory neurons (Fig. 8c, e). These findings indicate a sex-biased synaptic vulnerability, potentially reflecting both intrinsic neuronal dysfunction and secondary effects of altered glial support. Cell-cell communication networks are rewired in sex-distinct patterns in AD To delineate intercellular signaling alterations, ligand-receptor interaction analysis was performed across all cell types. Communication strength matrices revealed extensive network reorganization in AD, with markedly sex-specific patterns (Fig. 10a-d). Female AD brains exhibited globally elevated communication strength, particularly involving astrocytes and OPCs, compared to male AD and control groups (Fig. 10e-i). Differential analysis showed that female AD was characterized by weakened signaling from excitatory neurons to mature oligodendrocytes alongside strengthened communication toward OPCs, suggesting a redirection of neuronal support. In contrast, male AD displayed a more generalized but less intense enhancement of intercellular communication, with glial populations being particularly affected (Fig. 10j-l). Although ligand–receptor inference remains hypothesis-generating and does not confirm physical interaction, the consistent sex-specific rewiring detected across several cell-type pairs supports a biologically meaningful remodeling of intercellular signaling in AD. Overall, these multi-layered analyses converge on a model in which female AD is dominated by intrinsic metabolic and synaptic failure across neurons and glia, whereas male AD is characterized by coordinated neuroimmune activation. This positions biological sex as a fundamental axis of cellular vulnerability and network reorganization in Alzheimer’s disease. Discussion In this study, we provide a comprehensive, sex-stratified single-nucleus transcriptomic analysis of the human prefrontal cortex reveals a fundamental divergence in AD pathophysiology between males and females. Across neurons and glial populations, female AD brains exhibit a coordinated breakdown of metabolic, transcriptional, and synaptic homeostasis, whereas male AD brains display a distinct activation of neuroimmune and inflammatory pathways. These findings advance biological sex from a demographic descriptor to a primary determinant of cell-type vulnerability and intercellular network remodeling in AD. Sexually divergent pathological axes: homeostatic failure in females versus immune activation in males Our analysis demonstrates minimal overlap between male- and female-specific DEGs across all major brain cell types, indicating that AD does not represent a single molecular disease entity but rather two sex-polarized pathological states. In females, molecular alterations converge on core homeostatic systems-mitochondrial electron transport and oxidative phosphorylation in microglia, RNA splicing and microtubule dynamics in oligodendrocytes, and synaptic vesicle cycling in excitatory neurons. This convergence strongly suggests that energetic fragility and synaptic maintenance failure constitute central pillars of female-specific vulnerability. This interpretation is strongly supported by neuroimaging studies demonstrating pronounced cerebral glucose hypometabolism in women with AD [17, 18], reflecting impaired brain energy metabolism. Such vulnerability has been mechanistically linked to the abrupt decline in estrogen during menopause, a hormone critical for regulating mitochondrial function and glucose utilization [19, 20]. The marked downregulation of presynaptic genes in female excitatory neurons we observed provides a plausible cellular mechanism for the faster cognitive decline reported in women with AD [5]. Thus, our single-cell atlas reveals specific cellular and transcriptional mechanisms that align with established sex differences in AD clinical progression. In contrast, male AD brains displayed consistent enrichment of immune-related transcriptional programs, including cytokine production, lymphocyte activation, and other inflammatory processes-particularly within microglia and astrocytes. This immune signature corroborates prior reports of elevated pro-inflammatory cytokines in the peripheral leukocytes [21] and cerebrospinal fluid [22, 23] of male AD patients. Although a recent TSPO-PET study suggesting a stronger Aβ-independent microglial response in women [24], TSPO availability reflects proliferation or activation rather than functional polarization. Our findings suggest that while microglia in women may be reactive, their transcriptional state is skewed toward metabolic insufficiency, whereas male microglia exhibit immune activation supported by relatively preserved metabolic capacity. This interpretation also reconciles earlier FDG-PET observations reporting greater hypometabolism in men at comparable clinical stages of mild AD [25], suggesting stage-dependent shifts in sex differences. Disrupted estrogen signaling coordination in female AD Given the central role of estrogen in supporting neuronal and glial metabolic health, we examined estrogen signaling and identified a pattern of impaired pathway coordination in female AD. Despite ESR1 expression was increased across multiple cell types, correlations between ESR1 and other pathway members were attenuated, and enrichment scores for early and late estrogen-response gene sets were reduced. This pattern is consistent with models which receptor upregulation reflects a compensatory response to weakened signaling fidelity, rather than functional enhancement. Such paradoxical ESR1 elevation has been reported previously. For example, aging and AD are associated with increased estrogen receptor alpha (ERα/ESR1) expression accompanied by inactive splice isoforms [26], and declining ERα function can require higher ligand levels or alternative signaling mechanisms to maintain transcriptional output [27]. The disrupted coordination we observed aligns with sex-specific transcriptional rewiring reported in AD brain tissue [28, 29]. These molecular insights offer a biologically grounded interpretation of why late-initiated hormone replacement therapy (HRT) fails to confer cognitive benefit and amy even increase dementia risk [7, 8]. If estrogen signaling becomes structurally decoupled within the AD transcriptome, ligand supplementation alone may not restore neuroprotective activity. Astrocytic APOE downregulation connects genetic risk to female-specific metabolic-immune vulnerability The APOE ε4 allele exerts a stronger disease risk effect in women, a cornerstone epidemiologic finding of AD sexual dimorphism [8]. Our analysis reveals a potential molecular mechanism: APOE expression was significantly reduced specifically in female AD astrocytes, accompanied by extensive co-dysregulation of metabolic, oxidative stress, and inflammatory pathways. As astrocytes are the principal producers of ApoE in the brain, such reductions may compromise lipid homeostasis, metabolic support, and neuroimmune regulation. This finding integrates multiple lines of prior findings. Neuroimaging studies show stronger APOE ε4 by sex interactions on brain metabolism and structure decline in women [30, 31]. Experimental work similarly demonstrates sex-dependent astrocytic vulnerability: primary astrocytes from female APOE4-targeted replacement mice exhibit higher inflammatory gene expression [32], and female APOE4 mice show deficient astrocytic engagement with amyloid plaques and more extensive neurite injury [33]. Our human transcriptomic data extend these findings by situating astrocytic APOE downregulation at the nexus of a broader metabolic-immune dysregulation network, providing a single mechanistic framework for a unified explanation for sex-specific genetic risk amplification. Limitations and future directions This study has several limitations. First, analyses are cross-sectional and based on postmortem tissue, limiting causal inferences. Second, the dataset concentrates on the prefrontal cortex, and it remains uncertain whether similar sex-specific architectures exist across other AD-vulnerable regions. Third, snRNA-seq cannot differentiate true changes in cell abundance from transcriptional shifts affecting cluster identity. Fourth, endogenous estrogen levels were not available, precluding direct ligand-receptor correlation analyses. Finally, although relatively large for human single-cell datasets, the cohort size warrants validation in larger independent studies. Future work should functionally test whether restoring estrogen pathway coordination or astrocytic APOE expression can ameliorate female-specific metabolic and synaptic vulnerabilities. Longitudinal single-cell and spatial transcriptomic studies are needed to determine when these sex-divergent trajectories first emerge. More broadly, the distinct molecular signatures identified here—homeostatic failure in females and immune activation in males—should be evaluated as potential biomarkers for sex-informed patient stratification and as targets for personalized therapy strategies. Conclusion In conclusion, this study provides a comprehensive sex-stratified single-nucleus transcriptomic atlas of the human prefrontal cortex in Alzheimer's disease. Our findings demonstrate that biological sex fundamentally shapes the cellular and molecular landscape of AD. We identify two distinct pathological trajectories. In females, the disease is characterized by a widespread homeostatic collapse across metabolic, transcriptional, and synaptic domains. In males, it is dominated by sustained neuroimmune activation, while core metabolic functions remain relatively preserved. We further uncover a functional uncoupling of estrogen signaling in the female AD brain. Additionally, we link female-specific astrocytic APOE downregulation to the amplified genetic risk conferred by the APOE ε4 allele. Collectively, these findings position biological sex as a primary determinant of cellular vulnerability in AD. They strongly advocate for integrating sex-stratified frameworks into the development of precision diagnostics and targeted therapies. Methods Data acquisition and cohort curation Single-nucleus RNA sequencing (snRNA-seq) data from human prefrontal cortex were obtained from the integrated, pre-processed dataset published by Li et al. [13], which harmonized three foundational AD cohorts [14-16]. Age metadata were manually retrieved from the original publications and appended to the integrated dataset. The curated cohort included 75 individuals (37 AD, 38 control) with complete annotation for age, sex, disease status, neuropathology (Braak, CERAD), and APOE genotype, encompassing a total of 179,690 high-quality nuclei (Fig. 1). All analyses in this study were conducted exclusively on these harmonized human datasets. Pre-processing, quality control, and cell-type annotation We used the pre-integrated Seurat object from Li et al. without repeating batch correction to preserve consistency with the published harmonization workflow. As to maximize sensitivity for differential expression (DE) analysis, genes expressed in fewer than 3 nuclei across the dataset were filtered out. Major brain cell types were defined according to the original publication and validated using canonical marker genes: astrocytes (Ast), excitatory neurons (ex.neu; marker SLC17A7 ) [34, 35], inhibitory neurons (in.neu), microglia (Mic; marker CSF1R ) [36], oligodendrocytes (Oli; marker PLP1 ) [37], and oligodendrocyte precursor cells (OPC). All annotations were confirmed via marker-based heatmaps and cluster inspection (Fig. 2h). Differential expression analyses Sex-stratified DE analysis were conducted within each cell type using Seurat’s FindMarkers function with default setting. Comparisons were made between AD and CTL samples within each sex, yielding male-specific and female-specific DEGs. Differentially expressed genes (DEGs) were defined by an absolute log₂ fold-change (avg_log2FC) > 0.58 (≈1.5-fold) and an adjusted p-value (p_val_adj) < 0.05 (Benjamini-Hochberg correction). DEGs were visualized using volcano plots and summarized by Venn diagrams to quantify sex specificity (Fig. 3). Functional enrichment analysis Gene Ontology (GO) and KEGG pathway enrichment analyses were performed using clusterProfiler with default background gene sets. Sex-specific DEG lists were analyzed separately for each cell type. Sex-specific DEGs lists were analyzed separately for each cell type. Gene Set Enrichment Analysis (GSEA) was performed for pathways correlated with APOE expression in female AD astrocytes. A ranked gene list was generated based on Spearman correlation coefficients with APOE , and enrichment was tested against the HALLMARK gene sets from the msigdbr package using default parameters (Fig. 6c, d). Focused molecular and cellular analyses Key gene expression profiling : A curated panel of AD-relevant genes (e.g., TREM2 , TYROBP , APOE , C1QA , IL1B ) was selected based on prior human genetics and functional studies [38-40]. Sex-stratified log2 fold-change values were quantified across all major cell types (Fig. 4). Microglial functional module scoring: Three functional domains were defined using curated gene sets: i) metabolic activity (oxidative phosphorylation genes from KEGG pathway hsa00190) [41], ii) clearance capacity (phagocytic receptors, e.g., TREM2 , TYROBP ) [42, 43], and iii) immune activation (inflammatory cytokines and inflammasome components, e.g., IL1B , NLRP3 ) [42, 44]. Module scores were calculated using Seurat’s AddModuleScore, representing the average normalized expression across genes within each pathway (Fig. 7a). Estrogen signaling analysis: A core estrogen signaling gene set was curated to capture canonical receptor signaling and downstream neuroprotective programs, including: receptors ( ESR1 , ESR2 , PGR ), primary transcriptional target ( GREB1 ), and downstream effectors involved in neuroprotection ( BDNF , SIRT1 , BCL2 ), cell proliferation ( CCND1 , MYC ), and neuronal function ( NRCAM , VEGFA ) [45-47]. Pairwise Spearman correlations among these genes were computed. Gene Set Variation Analysis (GSVA) was employed to calculate enrichment scores for early and late estrogen response pathways across cell types (Fig. 5). Synaptic gene analysis in excitatory neurons: Synaptic gene sets were defined based on established synaptic biology literature [48-51] for: i) postsynaptic density (e.g., DLG1-4, GRIN1/2A/2B, SHANK1-3 ), ii) presynaptic active zone (e.g., SNAP25 , SYT1 , VAMP2 , NRXN1-3 ), and iii) synaptic vesicle components (e.g., SV2A-C, SYT1-5 ). Eight synaptic DEGs from female excitatory neurons were identified and combined with four biologically important synaptic genes, yielding 12 focal genes for sex-specific synaptic profiling (Fig. 8). Cell-cell communication analysis: Intercellular signaling networks were inferred using a custom workflow based on the CellChatDB.human ligand-receptor database. For each experimental group (F_AD, F_CTL, M_AD, M_CTL), the average expression of each ligand and receptor was calculated per cell type. The communication strength from cell type A to B was calculated as the sum of the products (average ligand expression in A) × (average receptor expression in B) across all relevant ligand-receptor pairs in database. This method generates probabilistic communication networks rather than validated physical interactions. Network-level comparisons and differential signaling matrices were constructed as shown in Fig. 9. Statistical analysis Differences in cell type proportions were assessed using two-sample z-tests for proportions, with Benjamini-Hochberg (BH) correction. Group comparisons of continuous variables (e.g., functional scores, gene expression) were performed using two-sided Wilcoxon rank-sum tests. For correlation analyses, Spearman's rank correlation was used for gene-gene correlations and Pearson's correlation for functional module associations. Multiple testing correction was applied where appropriate using the BH method. All statistical analyses were performed in R (v4.3.0). Software Analysis was conducted in R (v4.3.0). Primary analysis and related visualizations mainly utilized the following packages: Seurat (v5.0.0) for single-nucleus data handling; clusterProfiler and enrichplot for functional enrichment analysis; GSVA for gene set variation analysis; msigdbr for curated gene set collections; ggplot2 , ggpubr , and ComplexHeatmap for figure generation. A comprehensive list of dependencies is documented within the associated analysis code. Abbreviations Alzheimer’s disease (AD) Apolipoprotein E ε4 (APOE ε4) Hormone replacement therapy (HRT) Women’s Health Initiative Memory Study (WHIMS) Mild cognitive impairment (MCI) Estrogen receptor alpha (ESR1) SIngle-nucleus RNA sequencing (snRNA-seq) Excitatory neurons (ExN) Inhibitory neurons (InN) Astrocytes (Ast) Microglia (Mic) Oligodendrocytes (Oli) Oligodendrocyte precursor cells (OPC) Uniform manifold approximation and projection (UMAP) Differentially expressed genes (DEGs) Uniform Manifold Approximation and Projection (UMAP) Gene Ontology (GO) Gene Set Variation Analysis (GSVA) Gene set enrichment analysis (GSEA) Hormone replacement therapy (HRT) Differential expression (DE) Benjamini-Hochberg (BH) Declarations Ethics approval and consent to participate Not applicable. This study reanalyzed previously published, de-identified human postmortem brain snRNA-seq datasets. Ethical approvals and consent acquisition procedures were obtained by the original studies as reported in their respective publications. No new human or animal samples were collected for the present work. Consent for publication Not applicable. Code availability Custom R scripts used for analysis and figure generation are available from the corresponding author upon reasonable request. Data availability The snRNA-seq dataset analyzed in this study was obtained from the integrated, preprocessed resource published by Li et al. (2025). All underlying cohort data are available from the original publications and associated repositories. Accession identifiers and download instructions are provided in the Li et al. study and the source cohort papers. Competing interests The authors declare no conflict of interest. Funding This research received no external funding. Author Contributions Ziyi Zhou was responsible for the research design, methodology development, formal analysis, data collection and management, as well as drafting the initial manuscript. Chenxi Jiang contributed to the organization of literature review, refined analysis methods, and revised the manuscript. Zhenzhen Chen contributed to the manuscript’s revision, provided oversight and guidance throughout the research process, and managed the overall project. All authors have reviewed and approved the final version for publication. Acknowledgements Not applicable. References Livingston G, Huntley J, Sommerlad A, Ames D, Ballard C, Banerjee S, et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. The lancet. 2020;396(10248):413-46. Gauthier S, Albert M, Fox N, Goedert M, Kivipelto M, Mestre-Ferrandiz J, et al. Why has therapy development for dementia failed in the last two decades? Alzheimer's & Dementia. 2016;12(1):60-4. Ferretti MT, Iulita MF, Cavedo E, Chiesa PA, Schumacher Dimech A, Santuccione Chadha A, et al. Sex differences in Alzheimer disease—the gateway to precision medicine. Nature Reviews Neurology. 2018;14(8):457-69. Reitz C, Brayne C, Mayeux R. Epidemiology of Alzheimer disease. Nature Reviews Neurology. 2011;7(3):137-52. Lin KA, Choudhury KR, Rathakrishnan BG, Marks DM, Petrella JR, Doraiswamy PM, et al. Marked gender differences in progression of mild cognitive impairment over 8 years. Alzheimer's & dementia: translational research & clinical interventions. 2015;1(2):103-10. Neu SC, Pa J, Kukull W, Beekly D, Kuzma A, Gangadharan P, et al. Apolipoprotein E genotype and sex risk factors for Alzheimer disease: a meta-analysis. JAMA neurology. 2017;74(10):1178-89. Shumaker SA, Legault C, Rapp SR, Thal L, Wallace RB, Ockene JK, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. Jama. 2003;289(20):2651-62. Shumaker SA, Legault C, Kuller L, Rapp SR, Thal L, Lane DS, et al. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women's Health Initiative Memory Study. Jama. 2004;291(24):2947-58. Yanguas‐Casás N, Crespo‐Castrillo A, Arevalo MA, Garcia‐Segura LM. Aging and sex: Impact on microglia phagocytosis. Aging Cell. 2020;19(8):e13182. Mosconi L, Berti V, Dyke J, Schelbaum E, Jett S, Loughlin L, et al. Menopause impacts human brain structure, connectivity, energy metabolism, and amyloid-beta deposition. Scientific reports. 2021;11(1):10867. Jack Jr CR, Wiste HJ, Weigand SD, Knopman DS, Vemuri P, Mielke MM, et al. Age, sex, and APOE ε4 effects on memory, brain structure, and β-amyloid across the adult life span. JAMA neurology. 2015;72(5):511-9. Buckley RF, Mormino EC, Rabin JS, Hohman TJ, Landau S, Hanseeuw BJ, et al. Sex differences in the association of global amyloid and regional tau deposition measured by positron emission tomography in clinically normal older adults. JAMA neurology. 2019;76(5):542-51. Li Y, Serras CP, Blumenfeld J, Xie M, Hao Y, Deng E, et al. Cell-type-directed network-correcting combination therapy for Alzheimer’s disease. Cell. 2025;188(20):5516-34. e18. Mathys H, Davila-Velderrain J, Peng Z, Gao F, Mohammadi S, Young JZ, et al. Single-cell transcriptomic analysis of Alzheimer’s disease. Nature. 2019;570(7761):332-7. Zhou Y, Song WM, Andhey PS, Swain A, Levy T, Miller KR, et al. Human and mouse single-nucleus transcriptomics reveal TREM2-dependent and TREM2-independent cellular responses in Alzheimer’s disease. Nature medicine. 2020;26(1):131-42. Lau S-F, Cao H, Fu AK, Ip NY. Single-nucleus transcriptome analysis reveals dysregulation of angiogenic endothelial cells and neuroprotective glia in Alzheimer’s disease. Proceedings of the National Academy of Sciences. 2020;117(41):25800-9. Park J-C, Lim H, Byun MS, Yi D, Byeon G, Jung G, et al. Sex differences in the progression of glucose metabolism dysfunction in Alzheimer’s disease. Experimental & Molecular Medicine. 2023;55(5):1023-32. Nerattini M, Abenavoli EM, Caramelli F, Giacomucci G, Nacmias B, Mossello E, et al. Sex differences in brain glucose metabolism in alzheimer’s disease: A voxel-based study. GeroScience. 2025:1-11. Scheyer O, Rahman A, Hristov H, Berkowitz C, Isaacson R, Diaz Brinton R, et al. Female sex and Alzheimer’s risk: the menopause connection. The journal of prevention of Alzheimer's disease. 2018;5(4):225-30. Lopez-Lee C, Torres ERS, Carling G, Gan L. Mechanisms of sex differences in Alzheimer’s disease. Neuron. 2024;112(8):1208-21. Sochocka M, Ochnik M, Sobczyński M, Orzechowska B, Leszek J. Sex differences in innate immune response of peripheral blood leukocytes of Alzheimer’s disease patients. Archivum Immunologiae et Therapiae Experimentalis. 2022;70(1):16. Janelidze S, Mattsson N, Stomrud E, Lindberg O, Palmqvist S, Zetterberg H, et al. CSF biomarkers of neuroinflammation and cerebrovascular dysfunction in early Alzheimer disease. Neurology. 2018;91(9):e867-e77. Duarte-Guterman P, Albert AY, Inkster AM, Barha CK, Galea LA, Initiative AsDN. Inflammation in Alzheimer’s disease: do sex and APOE matter? Journal of Alzheimer’s Disease. 2020;78(2):627-41. Biechele G, Rauchmann B-S, Janowitz D, Buerger K, Franzmeier N, Weidinger E, et al. Associations between sex, body mass index and the individual microglial response in Alzheimer’s disease. Journal of neuroinflammation. 2024;21(1):30. Perneczky R, Drzezga A, Diehl-Schmid J, Li Y, Kurz A. Gender differences in brain reserve: an 18F-FDG PET study in Alzheimer's disease. Journal of neurology. 2007;254(10):1395-400. Ishunina TA, Fischer DF, Swaab DF. Estrogen receptor α and its splice variants in the hippocampus in aging and Alzheimer's disease. Neurobiology of aging. 2007;28(11):1670-81. Foster TC. Role of estrogen receptor alpha and beta expression and signaling on cognitive function during aging. Hippocampus. 2012;22(4):656-69. Sun LL, Yang SL, Sun H, Li WD, Duan SR. Molecular differences in Alzheimer's disease between male and female patients determined by integrative network analysis. Journal of Cellular and Molecular Medicine. 2019;23(1):47-58. Santiago JA, Quinn JP, Potashkin JA. Sex-specific transcriptional rewiring in the brain of Alzheimer’s disease patients. Frontiers in Aging Neuroscience. 2022;14:1009368. Sampedro F, Vilaplana E, De Leon MJ, Alcolea D, Pegueroles J, Montal V, et al. APOE-by-sex interactions on brain structure and metabolism in healthy elderly controls. Oncotarget. 2015;6(29):26663. Hohman TJ, Dumitrescu L, Barnes LL, Thambisetty M, Beecham G, Kunkle B, et al. Sex-specific association of apolipoprotein E with cerebrospinal fluid levels of tau. JAMA neurology. 2018;75(8):989-98. Mhatre-Winters I, Eid A, Han Y, Tieu K, Richardson JR. Sex and APOE genotype alter the basal and induced inflammatory states of primary astrocytes from humanized targeted replacement mice. ASN neuro. 2023;15(1):17590914221144549. Stephen T, Breningstall B, Suresh S, McGill C, Pike C. APOE genotype and biological sex regulate astroglial interactions with amyloid plaques in Alzheimer’s disease mice. Journal of neuroinflammation. 2022;19(1):286. Yao Z, Liu H, Xie F, Fischer S, Adkins RS, Aldridge AI, et al. A transcriptomic and epigenomic cell atlas of the mouse primary motor cortex. Nature. 2021;598(7879):103-10. Fremeau Jr RT, Kam K, Qureshi T, Johnson J, Copenhagen DR, Storm-Mathisen J, et al. Vesicular glutamate transporters 1 and 2 target to functionally distinct synaptic release sites. Science. 2004;304(5678):1815-9. Ginhoux F, Greter M, Leboeuf M, Nandi S, See P, Gokhan S, et al. Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science. 2010;330(6005):841-5. Klugmann M, Schwab MH, Pühlhofer A, Schneider A, Zimmermann F, Griffiths IR, et al. Assembly of CNS myelin in the absence of proteolipid protein. Neuron. 1997;18(1):59-70. Lue L-F, Schmitz C, Walker D. What happens to microglial TREM2 in Alzheimer’s disease: Immunoregulatory turned into immunopathogenic? Neuroscience. 2015;302:138-50. Haure-Mirande J-V, Audrain M, Ehrlich ME, Gandy S. Microglial TYROBP/DAP12 in Alzheimer’s disease: Transduction of physiological and pathological signals across TREM2. Molecular neurodegeneration. 2022;17(1):55. Villegas-Llerena C, Phillips A, Garcia-Reitboeck P, Hardy J, Pocock JM. Microglial genes regulating neuroinflammation in the progression of Alzheimer's disease. Current opinion in neurobiology. 2016;36:74-81. Kanehisa M, Furumichi M, Sato Y, Kawashima M, Ishiguro-Watanabe M. KEGG for taxonomy-based analysis of pathways and genomes. Nucleic acids research. 2023;51(D1):D587-D92. Keren-Shaul H, Spinrad A, Weiner A, Matcovitch-Natan O, Dvir-Szternfeld R, Ulland TK, et al. A unique microglia type associated with restricting development of Alzheimer’s disease. Cell. 2017;169(7):1276-90. e17. Ulland TK, Song WM, Huang SC-C, Ulrich JD, Sergushichev A, Beatty WL, et al. TREM2 maintains microglial metabolic fitness in Alzheimer’s disease. Cell. 2017;170(4):649-63. e13. Friedman BA, Srinivasan K, Ayalon G, Meilandt WJ, Lin H, Huntley MA, et al. Diverse brain myeloid expression profiles reveal distinct microglial activation states and aspects of Alzheimer’s disease not evident in mouse models. Cell reports. 2018;22(3):832-47. Toumba M, Kythreotis A, Panayiotou K, Skordis N. Estrogen receptor signaling and targets: Bones, breasts and brain. Molecular Medicine Reports. 2024;30(2):144. Marquardt RM, Kim TH, Shin J-H, Jeong J-W. Progesterone and estrogen signaling in the endometrium: what goes wrong in endometriosis? International journal of molecular sciences. 2019;20(15):3822. Park H, Pagan L, Tan O, Fadiel A, Demir N, Huang K, et al. Estradiol regulates expression of polysialated neural cell adhesion molecule by human vascular endothelial cells. Reproductive Sciences. 2010;17(12):1090-8. Südhof TC. Neurotransmitter release: the last millisecond in the life of a synaptic vesicle. Neuron. 2013;80(3):675-90. Fatemi SH, Eschenlauer A, Aman J, Folsom TD, Chekouo T. Quantitative proteomics of dorsolateral prefrontal cortex reveals an early pattern of synaptic dysmaturation in children with idiopathic autism. Cerebral Cortex. 2024;34(13):161-71. Gomez AM, Traunmüller L, Scheiffele P. Neurexins: molecular codes for shaping neuronal synapses. Nature Reviews Neuroscience. 2021;22(3):137-51. Bonnycastle K, Davenport EC, Cousin MA. Presynaptic dysfunction in neurodevelopmental disorders: Insights from the synaptic vesicle life cycle. Journal of Neurochemistry. 2021;157(2):179-207. Additional Declarations No competing interests reported. Supplementary Files SupplementaryTables.xlsx SupplementaryFigure.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9504119","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":628839566,"identity":"49a93169-56c9-4a93-8c2f-b4fc505dba64","order_by":0,"name":"Ziyi Zhou","email":"","orcid":"","institution":"Nankai University","correspondingAuthor":false,"prefix":"","firstName":"Ziyi","middleName":"","lastName":"Zhou","suffix":""},{"id":628839571,"identity":"9cc35c26-f132-46e0-a93c-c5d06e43fafc","order_by":1,"name":"Chenxi Jiang","email":"","orcid":"","institution":"Nankai University","correspondingAuthor":false,"prefix":"","firstName":"Chenxi","middleName":"","lastName":"Jiang","suffix":""},{"id":628839572,"identity":"45f8e9e2-f10c-4165-891e-19fc3a54db27","order_by":2,"name":"Zhenzhen Chen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2UlEQVRIiWNgGAWjYDACCSBmbACxeIC4IoGHVC1nSNbC2JZAWIf87OZjD3/usMljYD978HHhvDQZBv7DB/BqYZxzLN1A8kxaMQNPXrLxzG05PAwSafitYpbIMZMwbDuc2CDBYybNu60CqIXHAK8WNon8bxKJbf9BWsx/884BauE//wGvFh6JHDaJg20HwLYw8zYAHcaQg1cHg4REmplkY1tyYhtPjrE0z7E0HjaJNPwOk5+R/EzyZ5tdYj/7GcPPPDXJ9vz8hx/gtwbuKQzGKBgFo2AUjALyAQAN3jo5YDL0mgAAAABJRU5ErkJggg==","orcid":"","institution":"Karolinska Institutet","correspondingAuthor":true,"prefix":"","firstName":"Zhenzhen","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2026-04-23 08:25:31","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9504119/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9504119/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109185260,"identity":"c41d7f16-1b9c-4d52-b171-eb9c3a49edf3","added_by":"auto","created_at":"2026-05-13 10:59:49","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":312815,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eClinical Cohort Profiling and Neuropathological Characterization. (a)\u003c/strong\u003e Distribution of samples by diagnosis (AD, CTL) and sex. \u003cstrong\u003eb–c,\u003c/strong\u003e Counts of samples across \u003cstrong\u003e(b) \u003c/strong\u003eBraak stages and \u003cstrong\u003e(c) \u003c/strong\u003eCERAD scores;\u003cstrong\u003e (d)\u003c/strong\u003e Stratification of APOE genotypes by sex and clinical status; (\u003cstrong\u003ee–f)\u003c/strong\u003e Comparison of (e) Braak stage and (f) CERAD score between AD and CTL, showing significant pathological separation; \u003cstrong\u003e(g)\u003c/strong\u003e Age distribution across groups; t-test P-values indicate no significant age difference between sexes within groups; \u003cstrong\u003e(h–i)\u003c/strong\u003e Correlation analysis between age and (h) Braak stage or (i)\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/4d69b5eff4494cfef91ef016.png"},{"id":109205420,"identity":"89dac866-c1d4-42d9-9602-fceb521b200f","added_by":"auto","created_at":"2026-05-13 15:04:41","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":371206,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCellular atlas and sex-stratified cell-type composition in the integrated AD cohort\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea-e: \u003c/strong\u003eUniform Manifold Approximation and Projection (UMAP) visualization of 179,690 single-nucleus transcriptomes from the prefrontal cortex, integrated and clustered using Seurat and Harmony. \u003cstrong\u003e(a)\u003c/strong\u003eGlobal cell-type annotation reveals six major neural populations: astrocytes (Ast; n = 20,167), excitatory neurons (ex.neu; n = 56,973), inhibitory neurons (in.neu; n = 18,909), microglia (mic; n = 9,006), oligodendrocytes (oli; n = 61,784), and oligodendrocyte precursor cells (OPC; n = 12,851); \u003cstrong\u003eb-c,\u003c/strong\u003eUMAPs colored by biological sex \u003cstrong\u003e(b)\u003c/strong\u003e and diagnosis \u003cstrong\u003e(c)\u003c/strong\u003e; \u003cstrong\u003ed-e,\u003c/strong\u003eSex-resolved UMAPs for AD \u003cstrong\u003e(d)\u003c/strong\u003e and CTL \u003cstrong\u003e(e) \u003c/strong\u003esamples;\u003cstrong\u003e (f) \u003c/strong\u003eOverall cell-type composition across the dataset and within disease- and sex-stratified groups; (\u003cstrong\u003eg) \u003c/strong\u003eSex-stratified differences in cell-type proportions between AD and CTL groups. Bar height represents the difference in proportion (AD - CTL). Statistical significance was assessed using two-sample z-tests for proportions with Benjamini-Hochberg correction;\u003cstrong\u003e (h) \u003c/strong\u003eHeatmap showing average expression of canonical marker genes used for cell-type annotation\u003cstrong\u003e; (i)\u003c/strong\u003e Bubble plot summarizing the magnitude and direction of disease-associated proportion changes for each cell type by sex.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/846501720b243d650df802fc.png"},{"id":109185265,"identity":"cbd983e8-a45c-4a08-80c0-985a5a2b9d8a","added_by":"auto","created_at":"2026-05-13 10:59:49","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":328727,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSex-specific differential gene expression across major brain cell types. (a)\u003c/strong\u003e Heatmap summarizing the number of differentially expressed genes (DEGs) between AD and CTL groups, stratified by sex (Female/Male) and direction of change (Up/Down), across major brain cell types (ast, ex.neu, in.neu, mic, oli, opc) and overall; \u003cstrong\u003e(b) \u003c/strong\u003eBar chart comparing the total number of non-specific DEGs (red) with cell-type specific DEGs (blue) identified for each cell type; \u003cstrong\u003ec–h\u003c/strong\u003e Volcano plots displaying the versus for sex-specific DEGs in each cell type: (\u003cstrong\u003ec)\u003c/strong\u003e oligodendrocytes (oli);\u003cstrong\u003e (d)\u003c/strong\u003e excitatory neurons (ex.neu);\u003cstrong\u003e (e)\u003c/strong\u003e microglia (mic);\u003cstrong\u003e (f)\u003c/strong\u003e inhibitory neurons (in.neu); (\u003cstrong\u003eg)\u003c/strong\u003e OPCs; and\u003cstrong\u003e (h)\u003c/strong\u003e astrocytes (ast). For each cell type, results for Male (left) and Female (right) are shown separately. Significantly up-regulated (red) and down-regulated (blue) genes (FDR \u0026lt; 0.05) are highlighted.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/d43ef20cab0d2ede94e5efd3.png"},{"id":109185299,"identity":"8c8a00e1-125d-4e5d-a915-ee3e8786df0c","added_by":"auto","created_at":"2026-05-13 11:00:02","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":560504,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFunctional enrichment analysis of sex-specific DEGs. (a–f) \u003c/strong\u003eVenn diagrams showing the overlap between male (M) and female (F) differentially expressed genes (DEGs) for major brain cell types: \u003cstrong\u003e(a)\u003c/strong\u003eexcitatory neurons (ex.neu); (\u003cstrong\u003eb)\u003c/strong\u003e inhibitory neurons (in.neu);\u003cstrong\u003e (c)\u003c/strong\u003eastrocytes (ast); \u003cstrong\u003e(d)\u003c/strong\u003e oligodendrocytes (oli);\u003cstrong\u003e (e)\u003c/strong\u003e oligodendrocyte precursor cells (opc); and\u003cstrong\u003e (f)\u003c/strong\u003e microglia (mic); \u003cstrong\u003eg–l, \u003c/strong\u003eGene Ontology (GO) biological process enrichment analysis for sex-specific DEGs across various cell types. Bar plots show the top enriched terms for: \u003cstrong\u003e(g) \u003c/strong\u003efemale excitatory neurons (ex.neu); \u003cstrong\u003e(h) \u003c/strong\u003efemale oligodendrocytes (oli);\u003cstrong\u003e (i) \u003c/strong\u003efemale astrocytes (ast); \u003cstrong\u003e(j)\u003c/strong\u003e male astrocytes (ast); \u003cstrong\u003e(k)\u003c/strong\u003e female microglia (mic); and \u003cstrong\u003e(l)\u003c/strong\u003e male microglia (mic). Bar length corresponds to the −log₁₀ (adjusted p value), and the color gradient represents the gene count associated with each term.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/2225b1a04f4f7ef31bced5a8.png"},{"id":109185261,"identity":"7c7f11c6-95a6-4905-af0b-c800edc7243d","added_by":"auto","created_at":"2026-05-13 10:59:49","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":158830,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSex-stratified AD-associated gene expression signatures across cell types. (a) \u003c/strong\u003eHeatmap showing log₂ fold-change (AD vs CTL) for selected AD-associated genes across major cell types, analyzed separately in females and males.\u003cstrong\u003e (b) \u003c/strong\u003eBubble plot depicting sex-specific expression changes of the same gene set. Bubble size indicates mean expression in CTL samples, and color denotes log₂ fold change in AD.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/2bc55ea50136af9a117c7e28.png"},{"id":109185297,"identity":"4e39778f-72e4-4e93-8871-d094033720ab","added_by":"auto","created_at":"2026-05-13 10:59:59","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":195085,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDysregulated of estrogen signaling in female Alzheimer’s disease. (a)\u003c/strong\u003e Heatmaps of Spearman correlation coefficients among core estrogen signaling genes in female brain cells. Left, AD; right, CTL. All displayed correlations are significant after Benjamini–Hochberg correction; \u003cstrong\u003e(b) \u003c/strong\u003eBubble plot showing changes in correlation strength (Δρ = ρ_AD − ρ_CTL) between ESR1 and other estrogen pathway genes in females; \u003cstrong\u003e(c)\u003c/strong\u003e Scatter plot comparing \u003cem\u003eESR1\u003c/em\u003e-centered correlation coefficients in AD versus CTL samples, with color indicating Δρ; \u003cstrong\u003e(d)\u003c/strong\u003eGene Set Variation Analysis (GSVA) scores for early and late estrogen response programs across female cell types. Bars indicate mean differences (AD − CTL) with 95% confidence intervals;\u003cstrong\u003e (e)\u003c/strong\u003e Network diagram illustrating functional relationships among estrogen pathway genes; \u003cstrong\u003e(f)\u003c/strong\u003e GSVA scores for estrogen response pathways in male samples.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/0a4d4d44068046fa0c3235a1.png"},{"id":109185298,"identity":"ebacac40-6b49-401f-9aca-d7a3af87d0aa","added_by":"auto","created_at":"2026-05-13 11:00:01","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":260548,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAstrocytic APOE expression and associated molecular networks in female AD. (a)\u003c/strong\u003e Violin plots showing sex-stratified APOE expression across major cell types in AD and CTL samples; \u003cstrong\u003e(b)\u003c/strong\u003eDistribution of genome-wide Spearman correlation coefficients between APOE and all detected transcripts in female AD astrocytes.\u003cstrong\u003e (c)\u003c/strong\u003e Gene set enrichment analysis (GSEA) of APOE-associated genes in female AD astrocytes;\u003cstrong\u003e(d) \u003c/strong\u003eEnrichment plots for selected pathways highlighted in c, illustrating coordinated metabolic and immune dysregulation.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/2694406a16a5ed876748a510.png"},{"id":109185263,"identity":"c51bb710-5d16-41a7-bfc3-87d18a00a944","added_by":"auto","created_at":"2026-05-13 10:59:49","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":275099,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSex-dependent microglial functional alterations in AD. (a)\u003c/strong\u003e Violin plots of microglial functional module scores for metabolic activity, clearance capacity, and immune activation, stratified by sex and diagnosis; \u003cstrong\u003e(b)\u003c/strong\u003e Expression distributions of key microglial (\u003cem\u003eP2RY12\u003c/em\u003e, \u003cem\u003eCX3CR1\u003c/em\u003e, \u003cem\u003eTREM2\u003c/em\u003e, \u003cem\u003eAPOE\u003c/em\u003e) across sex-diagnosis groups; \u003cstrong\u003e(c) \u003c/strong\u003eCorrelation between metabolic and clearance module scores in female AD microglia.\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/914047d72c113eab87335b06.png"},{"id":109185295,"identity":"4178c68f-9a06-44a5-bb0f-682a6ae270e9","added_by":"auto","created_at":"2026-05-13 10:59:57","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":364946,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSex-biased synaptic gene dysregulation in excitatory neurons.\u003c/strong\u003e\u003cbr\u003e\n \u003cstrong\u003e(a)\u003c/strong\u003e Expression changes of selected synaptic genes in AD versus CTL samples; \u003cstrong\u003e(b)\u003c/strong\u003e Average expression changes for presynaptic, postsynaptic, and synaptic adhesion gene categories; \u003cstrong\u003e(c)\u003c/strong\u003e Heatmap of synaptic gene expression across samples grouped by sex and diagnosis; (\u003cstrong\u003ed)\u003c/strong\u003e Heatmap of sex-stratified log₂ fold changes (AD vs. CTL); (\u003cstrong\u003ee)\u003c/strong\u003e Violin plots illustrating expression distributions of key synaptic genes across sex-diagnosis groups;\u003cstrong\u003e (f)\u003c/strong\u003e Line-point plot summarizing functional category-level expression trends by sex.\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/a4c401b6e10621a5ca3d26c2.png"},{"id":109205410,"identity":"b0131d69-2052-4351-b324-b86cf90c091a","added_by":"auto","created_at":"2026-05-13 15:04:38","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":502104,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSex-distinct remodeling inferred cell–cell communication networks in AD. (a–d) \u003c/strong\u003eHeatmaps depicting inferred cell-cell communication strengths for female AD (a), female CTL (b), male AD (c), and male CTL (d) groups.\u003cstrong\u003e (e)\u003c/strong\u003e Integrated comparison and clustering of communication networks across all groups;\u003cstrong\u003e (f–i)\u003c/strong\u003e Chord diagrams illustrating intercellular communication patterns in each group; \u003cstrong\u003e(j–l)\u003c/strong\u003e Differential communication heatmaps comparing female AD vs. CTL (j), male AD vs. CTL, and (k), and female AD vs. male AD (l).\u003c/p\u003e","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/f2862254ba615dd2802d4a54.png"},{"id":109405543,"identity":"5a9181c8-f07e-4a16-8f61-cc8d7b6f3368","added_by":"auto","created_at":"2026-05-17 13:18:52","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3624638,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/55e1d26c-750b-4399-be7f-76e44e81b0e7.pdf"},{"id":109185296,"identity":"15d50e6f-ad6e-4e1d-8060-6426072130b7","added_by":"auto","created_at":"2026-05-13 10:59:58","extension":"xlsx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":8995574,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTables.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/59b7b81ac269d5aaecdf9e63.xlsx"},{"id":109185302,"identity":"c5f4218e-df6a-4b86-b4c8-00bcf39192a6","added_by":"auto","created_at":"2026-05-13 11:00:06","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1920167,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigure.docx","url":"https://assets-eu.researchsquare.com/files/rs-9504119/v1/fcd53e195039d961f7705d9d.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Single-Nucleus Transcriptomics Analysis Identifies Sex-Dichotomous Pathological Axes in Alzheimer’s Disease: Female Homeostatic Failure versus Male Neuroimmune Activation","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAlzheimer\u0026rsquo;s disease (AD) is the most prevalent neurodegenerative disorder worldwide and constitutes a rapidly escalating public health burden in aging societies [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Therapeutic development has focused intensively on canonical pathological hallmarks, including beta-amyloid plaques and intracellular neurofibrillary tangles, yet these methods have gained limited clinical success. The majority of late-stage trials that target these pathways have failed to prove meaningful efficacy [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. One factor underlying this persistent translational impasse is the profound clinical and biological heterogeneity of AD. It has thus become urgent to identify the biological modifiers shaping this heterogeneity and to integrate them into precision medicine frameworks. Among such modifiers, biological sex has gained substantial recognition as a determinant of AD risk, progression, and molecular pathology [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eEpidemiological and clinical studies demonstrate pronounced sex differences in AD prevalence and disease trajectory. Women account for approximately two-thirds of sporadic AD cases, indicating a markedly higher lifetime risk compared to men [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Longitudinal studies demonstrate that women diagnosed with AD experience a more rapid rate of cognitive decline than their male counterparts [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Genetic susceptibility also exhibits sex-dependent effects. Meta-analyses have shown that female carriers of the apolipoprotein E ε4 (APOE ε4) allele, particularly between the ages of 65 and 75, face a significantly higher risk of developing AD than male carriers of the same genotype [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eMultiple biological mechanisms likely contribute to this disparity. At the genetic level, APOE ε4 exerts a stronger risk-amplifying effect in women within specific age windows, while the potentially protective APOE ε2 allele may confer greater benefit in women than in men. Endocrine factors also play a pivotal role. Women experience a relatively abrupt decline in circulating estrogen levels during the menopausal transition, in contrast to the more gradual age-related decline in androgens in men. Clinical investigations support the \u0026ldquo;critical window hypothesis\u0026rdquo; for hormone replacement therapy (HRT), suggesting that neuroprotective effects may only be achievable when therapy is initiated during the perimenopausal period. However, large randomized controlled trials, including the Women\u0026rsquo;s Health Initiative Memory Study (WHIMS), demonstrated that combined estrogen-plus-progestin therapy increases the risk of dementia and mild cognitive impairment (MCI) in postmenopausal women aged 65 years and older [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Estrogen-only therapy likewise failed to reduce dementia risk and may increase the combined incidence of dementia and MCI in this population [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], leading to recommendations against HRT for dementia prevention.\u003c/p\u003e \u003cp\u003eNeuroimmune mechanisms, particularly those involving microglia, further contribute to sex-specific vulnerability in AD. Experimental studies indicate that microglia from aged female models exhibit enhanced basal phagocytic capacity for neuronal debris but display impaired dynamic regulation of this function in response to inflammatory stimuli [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. In human postmortem brain tissue, estrogen receptor alpha (ESR1) expression positively correlates with the microglial phagocytosis-associated receptor TREM2 in samples from postmenopausal women, suggesting a potential link between estrogen signaling and microglial function in the aging female brain [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Consistent with these molecular findings, neuroimaging studies reveal that women may exhibit greater hippocampal atrophy at Aβ burden levels comparable to those at which men remain relatively stable [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Furthermore, women exhibit stronger associations between global Aβ burden, regional tau pathology, and cognitive decline than men at similar levels of amyloid deposition [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eDespite these advances, major gaps remain in understanding the molecular and cellular mechanisms driving sex differences in AD. Current research examining interactions among estrogen signaling, APOE genotype, and neuroinflammatory pathways remains fragmented, and their combined or cell-type-specific effects are poorly resolved [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Critically, there is a lack of systematic analysis characterizing sex-specific transcriptional programs and functional states across diverse brain cell types (e.g., neurons, astrocytes, microglia) at single-cell resolution in human AD brains [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Furthermore, commonly used animal models inadequately capture the prolonged human perimenopausal transition and the complex temporal dynamics of brain aging, limiting their translational relevance for studying sex-dependent mechanisms [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTo address these limitations, systematic bioinformatic analyses leveraging high-quality human single-cell datasets stratified by sex are essential for elucidating the cellular and molecular basis of sexual dimorphism in AD. In this study, we analyzed an integrated human single-nucleus RNA sequencing (snRNA-seq) dataset published by Li et al. (2025) in Cell [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], which harmonizes data from three foundational AD cohorts Mathys et al. (2019) [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], Zhou et al. (2020) [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], and Lau et al. (2020) [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] using rigorous cross-platform normalization and batch correction. This dataset provides a robust framework for cell-type resolved investigation of AD-associated transcriptional changes. Unlike the original study, we explicitly model AD transcriptional programs through a sex-stratified and sex-by-diagnosis framework across major brain cell types, integrating differential expression, pathway coordination, functional module scoring, and network-level signaling inference to delineate sex-dependent cellular states.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eIntegrated cohort validation and confounder control\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo systematically investigate sex-stratified cellular pathology in AD, we leveraged the pre-processed, harmonized single-nucleus RNA sequencing (snRNA-seq) dataset of the human prefrontal cortex published by Li et al. (2025) [13], which integrates three foundational AD cohorts [14-16]. We curated the dataset by retrieving and appending complete age metadata manually from the original studies, achieving a final cohort of 75 samples (37 AD, 38 control [CTL]) with 100% completeness for key clinical variables.\u003c/p\u003e\n\u003cp\u003eCERAD score, demonstrating that age does not confound pathological severity in this dataset.\u003c/p\u003e\n\u003cp\u003eSystematic cohort characterization validated robust data, which were consistent with expected neuropathological features (Fig. 1). As anticipated, AD cases exhibited significantly higher Braak stages (Fig. 1b, e) and lower CERAD scores (Fig. 1c, f) compared to controls, validating diagnostic annotation. Within each sex, age distribution did not differ significantly between AD and CTL groups (Fig. 1g-i), and APOE \u0026epsilon;4 allele frequency followed established disease-risk patterns (Fig. 1d). No significant sex-by-age interaction was detected within either diagnostic group, supporting the absence of major demographic confounding. Together, these results establish a well-balanced cohort that provides a solid foundation for downstream sex-stratified, cell-type-specific analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCellular atlas reveals sex-dimorphic shifts in cell-type proportions in AD\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe analyzed 179,690 high-quality nuclei from the integrated dataset and resolved six major neural cell populations (Fig. 2a). For clarity, we refer to these as excitatory neurons (ExN), inhibitory neurons (InN), astrocytes (Ast), microglia (Mic), oligodendrocytes (Oli), and oligodendrocyte precursor cells (OPC); corresponding labels in figures are ex.neu, in.neu, ast, mic, oli, and opc, respectively. Uniform manifold approximation and projection (UMAP) visualization demonstrated effective integration across source cohorts, sexes (Fig. 2b), and diagnostic groups (Fig. 2c), with no gross spatial segregation observed in sex-resolved AD or control subsets (Fig. 2d, e). Cell-type annotations were further validated by canonical marker gene expression (Fig. 2h).\u003c/p\u003e\n\u003cp\u003eComparative analysis of cell-type proportions uncovered subtle yet statistically significant alterations in AD that diverged by sex (Fig. 2f-g, i). In both sexes, AD was associated with an increased proportion of astrocytes and a reduction in inhibitory neurons. Strikingly, excitatory neurons and oligodendrocytes exhibited opposing trajectories: female AD samples showed an increase in the proportion of excitatory neurons accompanied by a decrease in oligodendrocytes, whereas male AD samples displayed the inverse pattern. Though such changes may reflect selective cellular vulnerability or survival, transcriptional state changes influencing cell-type assignment cannot be fully excluded and are addressed in subsequent functional analyses. Collectively, these findings suggest that AD engages distinct cellular remodeling programs in males and females.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTranscriptomic profiling uncovers sex-polarized functional dysregulation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo sensitively capture molecular alterations, gene filtering thresholds were recalibrated to retain genes expressed in at least 3 cells, maximizing detection power for differential expression analysis (see Methods). Cell-type-stratified analysis between AD and CTL samples was performed separately for each sex. Across all cell types, differential expression analysis revealed widespread transcriptional alterations exhibiting minimal cross-sex overlap among differentially expressed genes (DEGs) (Fig. 3a, b). Venn diagrams and volcano plots for each cell type quantified this divergence, demonstrating that the majority of DEGs were sex-specific (Fig. 3c-h; Fig. 4a-f). Importantly, this sex specificity was not driven by differences in cell number or sequencing depth, supporting a biological rather than technical origin of the observed polarization.\u003c/p\u003e\n\u003cp\u003eGene Ontology (GO) enrichment analysis of female-specific DEGs revealed a coherent pattern of disruption in pathways critical for intracellular homeostasis and neuronal functions (Fig. 4g, h, i, k). In females, microglial DEGs were overwhelmingly enriched for mitochondrial electron transport and oxidative phosphorylation. Oligodendrocyte DEGs were associated with RNA splicing and microtubule polymerization, while excitatory neuron DEGs were enriched for synaptic vesicle cycling, neurotransmitter secretion, and axonogenesis. Together, these signatures point to a concerted impairment of bioenergetic, post-transcriptional regulatory, and synaptic machinery in the female AD brain.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn contrast, GO analysis of male-specific DEGs highlighted a distinct pathological axis centered on immune activation and inflammatory signaling (Fig. 4j, l). Male microglial and astrocyte DEGs were significantly enriched for pathways related to T cell differentiation, lymphocyte activation, and cytokine production, indicating a pronounced neuroinflammatory response. These sex-divergent functional programs were further supported by complementary KEGG pathway analysis (Supplementary Fig. S1).\u003c/p\u003e\n\u003cp\u003eFocused examination of high-priority genes implicated in AD genetics and the identified functional signatures, including immune regulators (e.g., \u003cem\u003eTREM2\u003c/em\u003e, \u003cem\u003eTYROBP\u003c/em\u003e), the major genetic risk factor \u003cem\u003eAPOE\u003c/em\u003e, synaptic genes (e.g., \u003cem\u003eSNAP25\u003c/em\u003e, \u003cem\u003eSYT1\u003c/em\u003e), and the estrogen receptor \u003cem\u003eESR1\u003c/em\u003e. Their cell-type-resolved expression profiles confirmed the profound sex-specificity of AD-associated transcriptional alterations (Fig. 5a, b). In females, AD was characterized by a pronounced downregulation of synaptic genes, particularly \u003cem\u003eSNAP25\u003c/em\u003e and \u003cem\u003eSYT1\u003c/em\u003e, most evident in oligodendrocytes and oligodendrocyte precursor cells. \u003cem\u003eAPOE\u003c/em\u003e expression decreased in astrocytes but increased in microglia, while \u003cem\u003eESR1\u003c/em\u003e expression was elevated across multiple cell types. In males, changes in these genes were comparatively attenuated. Detailed expression statistics are provided in Supplementary Table 3.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEstrogen signaling pathway is dysregulated in female AD\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGiven the pronounced female-specific transcriptional alterations, we investigated estrogen signaling using a curated gene set encompassing nuclear receptors (ESR1, ESR2, PGR), primary transcriptional targets (GREB1), and downstream effectors involved in neuroprotection (BDNF, SIRT1, BCL2), cell proliferation (CCND1, MYC), and neuronal function (NRCAM, VEGFA) (Fig. 6e).\u003c/p\u003e\n\u003cp\u003ePathway coordination was first assessed via pairwise Spearman correlations among estrogen pathway genes in female brain cells. Consistent with prior single-cell studies, absolute correlation coefficients were modest; however, pathway-level coordination was statistically robust in both AD and control groups (all adjusted p \u0026lt; 0.05; Fig. 6a). Differential correlation analysis revealed substantial rewiring of ESR1-centered interactions in AD (Fig. 6b, c). Correlations between ESR1 and several pathway components were attenuated (e.g., ESR2, SIRT1), whereas the association with the cell cycle regulator CCND1 was strengthened. This occurred alongside a generalized increase in ESR1 expression across female AD cell types (Fig. 5), suggesting compensatory upregulation amid impaired pathway coordination.\u003c/p\u003e\n\u003cp\u003eFunctional validation using Gene Set Variation Analysis (GSVA) demonstrated broad dampening of early and late estrogen response programs in female AD brains, most prominently in astrocytes and OPCs (Fig. 6d). In contrast, male AD samples exhibited minimal and inconsistent changes in these pathways (Fig. 6f). Together, these findings delineate a distinct impairment of estrogen signaling networks in female AD, characterized by discordant gene\u0026ndash;gene associations and lowered\u003cem\u003e\u0026nbsp;\u003c/em\u003efunctional output despite elevated ESR1 expression.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAstrocytic APOE reduction is linked to metabolic-immune dysregulation in female AD\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGiven the central role of \u003cem\u003eAPOE\u003c/em\u003e as the strongest genetic risk factor for AD, we further dissected its sex- and cell-type-specific contributions. \u003cem\u003eAPOE\u003c/em\u003e expression was predominantly localized to astrocytes, with lower expression in microglia and minimal expression in neurons (Fig. 7a). A significant reduction in \u003cem\u003eAPOE\u003c/em\u003e expression was observed specifically in female AD astrocytes, accompanied by increased expression in female microglia. Male samples showed similar but notably attenuated trends.\u003c/p\u003e\n\u003cp\u003eGenome-wide correlation analysis in female astrocytes revealed extensive co-dysregulation associated with \u003cem\u003eAPOE\u003c/em\u003e expression (Fig. 7b). Gene set enrichment analysis (GSEA) revealed significant enrichment of pathways related to energy metabolism, lipid handling, oxidative stress, complement activation, and inflammatory signaling (all FDR \u0026lt; 0.001; Fig. 7c, d). While causality cannot be inferred from correlation-based analyses, the strong coupling of astrocytic APOE reduction with metabolic and immune pathways suggests a central role for astrocytic dysfunction in female AD vulnerability. The concomitant increase of APOE in microglia may reflect compensatory redistribution of functional burden across glial compartments.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMicroglial functional states exhibit sex-dichotomous dysregulation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo characterize microglial functional alterations, module scores were calculated for metabolic activity, clearance capacity, and immune activation (Fig. 8a). Female AD microglia displayed marked metabolic impairment compared to controls (p \u0026lt; 0.0001), accompanied by a modest but significant reduction in clearance capacity (p \u0026lt; 0.01). In contrast, male AD microglia showed preserved metabolic function but a pronounced increase in immune activation relative to male controls (p \u0026lt; 0.0001).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA weak but significant positive correlation between metabolic and clearance scores was detected specifically in female AD microglia (Pearson\u0026rsquo;s r = 0.10, p = 5.03 \u0026times; 10⁻⁵; Fig. 7c). Gene-level analysis supported these functional patterns: male AD microglia exhibited modest downregulation of the homeostatic marker P2RY12 and upregulation of TREM2, whereas female AD microglia showed elevated APOE expression (Fig. 8b). Together, these results reveal a sex-dependent inversion of microglial functional states in AD, with female microglia characterized by metabolic insufficiency and males by heightened inflammatory activation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExcitatory neurons exhibit sex-divergent synaptic gene dysregulation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMotivated by synaptic pathway enrichment in female excitatory neuron DEGs, we analyzed a focused panel of synaptic genes encompassing presynaptic vesicle release, postsynaptic transduction, and synaptic adhesion. In pooled AD samples, presynaptic genes \u003cem\u003e(SYP, SNAP25, RAB3A, VAMP2)\u003c/em\u003e were predominantly downregulated, whereas several postsynaptic genes (\u003cem\u003eGRIN1, DLG2\u003c/em\u003e) were upregulated (Fig. 9a, b).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSex-stratified analysis revealed that this presynaptic\u0026ndash;postsynaptic imbalance was substantially more pronounced in females. Female AD samples exhibited severe downregulation of vesicle release machinery, particularly VAMP2 and RAB3A, whereas alterations in male AD samples were comparatively mild (Fig. 9d, f). Expression distributions further highlighted increased heterogeneity and pronounced presynaptic loss specifically in female AD excitatory neurons (Fig. 8c, e). These findings indicate a sex-biased synaptic vulnerability, potentially reflecting both intrinsic neuronal dysfunction and secondary effects of altered glial support.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCell-cell communication networks are rewired in sex-distinct patterns in AD\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo delineate intercellular signaling alterations, ligand-receptor interaction analysis was performed across all cell types. Communication strength matrices revealed extensive network reorganization in AD, with markedly sex-specific patterns (Fig. 10a-d). Female AD brains exhibited globally elevated communication strength, particularly involving astrocytes and OPCs, compared to male AD and control groups (Fig. 10e-i).\u003c/p\u003e\n\u003cp\u003eDifferential analysis showed that female AD was characterized by weakened signaling from excitatory neurons to mature oligodendrocytes alongside strengthened communication toward OPCs, suggesting a redirection of neuronal support. In contrast, male AD displayed a more generalized but less intense enhancement of intercellular communication, with glial populations being particularly affected (Fig. 10j-l). Although ligand\u0026ndash;receptor inference remains\u0026nbsp;hypothesis-generating and does not confirm physical interaction, the consistent sex-specific rewiring\u0026nbsp;detected across several\u0026nbsp;cell-type pairs supports a biologically meaningful remodeling of intercellular signaling in AD.\u003c/p\u003e\n\u003cp\u003eOverall, these multi-layered analyses converge on a model in which female AD is dominated by intrinsic metabolic and synaptic failure across neurons and glia, whereas male AD is characterized by coordinated neuroimmune activation. This positions biological sex as a fundamental axis of cellular vulnerability and network reorganization in Alzheimer\u0026rsquo;s disease.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we provide a comprehensive, sex-stratified single-nucleus transcriptomic analysis of the human prefrontal cortex reveals a fundamental divergence in AD pathophysiology between males and females. Across neurons and glial populations, female AD brains exhibit a coordinated breakdown of metabolic, transcriptional, and synaptic homeostasis, whereas male AD brains display a distinct activation of neuroimmune and inflammatory pathways. These findings advance biological sex from a demographic descriptor to a primary determinant of cell-type vulnerability and intercellular network remodeling in AD.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSexually divergent pathological axes: homeostatic failure in females versus immune activation in males\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur analysis demonstrates minimal overlap between male- and female-specific DEGs across all major brain cell types, indicating that AD does not represent a single molecular disease entity but rather two sex-polarized pathological states. In females, molecular alterations converge on core homeostatic systems-mitochondrial electron transport and oxidative phosphorylation in microglia, RNA splicing and microtubule dynamics in oligodendrocytes, and synaptic vesicle cycling in excitatory neurons. This convergence strongly suggests that energetic fragility and synaptic maintenance failure constitute central pillars of female-specific vulnerability.\u003c/p\u003e\n\u003cp\u003eThis interpretation is strongly supported by neuroimaging studies demonstrating pronounced cerebral glucose hypometabolism in women with AD [17, 18], reflecting impaired brain energy metabolism. Such vulnerability has been mechanistically linked to the abrupt decline in estrogen during menopause, a hormone critical for regulating mitochondrial function and glucose utilization [19, 20]. The marked downregulation of presynaptic genes in female excitatory neurons we observed provides a plausible cellular mechanism for the faster cognitive decline reported in women with AD [5]. Thus, our single-cell atlas reveals specific cellular and transcriptional mechanisms that align with established sex differences in AD clinical progression.\u003c/p\u003e\n\u003cp\u003eIn contrast, male AD brains displayed consistent enrichment of immune-related transcriptional programs, including cytokine production, lymphocyte activation, and other inflammatory processes-particularly within microglia and astrocytes. This immune signature corroborates prior reports of elevated pro-inflammatory cytokines in the peripheral leukocytes [21] and cerebrospinal fluid [22, 23] of male AD patients. Although a recent TSPO-PET study suggesting a stronger Aβ-independent microglial response in women [24], TSPO availability reflects proliferation or activation rather than functional polarization. Our findings suggest that while microglia in women may be reactive, their transcriptional state is skewed toward metabolic insufficiency, whereas male microglia exhibit immune activation supported by relatively preserved metabolic capacity. This interpretation also reconciles earlier FDG-PET observations reporting greater hypometabolism in men at comparable clinical stages of mild AD [25], suggesting stage-dependent shifts in sex differences.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDisrupted estrogen signaling coordination in female AD\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGiven the central role of estrogen in supporting neuronal and glial metabolic health, we examined estrogen signaling and identified a pattern of impaired pathway coordination in female AD. Despite ESR1 expression was increased across multiple cell types, correlations between ESR1 and other pathway members were attenuated, and enrichment scores for early and late estrogen-response gene sets were reduced. This pattern is consistent with models which receptor upregulation reflects a compensatory response to weakened signaling fidelity, rather than functional enhancement.\u003c/p\u003e\n\u003cp\u003eSuch paradoxical ESR1 elevation has been reported previously. For example, aging and AD are associated with increased estrogen receptor alpha (ERα/ESR1) expression accompanied by inactive splice isoforms [26], and declining ERα function can require higher ligand levels or alternative signaling mechanisms to maintain transcriptional output [27]. The disrupted coordination we observed aligns with sex-specific transcriptional rewiring reported in AD brain tissue [28, 29]. These molecular insights offer a biologically grounded interpretation of why late-initiated hormone replacement therapy (HRT) fails to confer cognitive benefit and amy even increase dementia risk [7, 8]. If estrogen signaling becomes structurally decoupled within the AD transcriptome, ligand supplementation alone may not restore neuroprotective activity.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAstrocytic APOE downregulation connects genetic risk to female-specific metabolic-immune vulnerability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe APOE ε4 allele exerts a stronger disease risk effect in women, a cornerstone epidemiologic finding of AD sexual dimorphism [8]. Our analysis reveals a potential molecular mechanism: APOE expression was significantly reduced specifically in female AD astrocytes, accompanied by extensive co-dysregulation of metabolic, oxidative stress, and inflammatory pathways. As astrocytes are the principal producers of ApoE in the brain, such reductions may compromise lipid homeostasis, metabolic support, and neuroimmune regulation.\u003c/p\u003e\n\u003cp\u003eThis finding integrates multiple lines of prior findings. Neuroimaging studies show stronger APOE ε4 by sex interactions on brain metabolism and structure decline in women [30, 31]. Experimental work similarly demonstrates sex-dependent astrocytic vulnerability: primary astrocytes from female APOE4-targeted replacement mice exhibit higher inflammatory gene expression [32], and female APOE4 mice show deficient astrocytic engagement with amyloid plaques and more extensive neurite injury [33]. Our human transcriptomic data extend these findings by situating astrocytic APOE downregulation at the nexus of a broader metabolic-immune dysregulation network, providing a\u0026nbsp;single mechanistic framework\u0026nbsp;for a unified explanation for sex-specific genetic risk amplification.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLimitations and future directions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study has several limitations. First, analyses are cross-sectional and based on postmortem tissue, limiting causal inferences. Second, the dataset concentrates on the prefrontal cortex, and it remains uncertain whether similar sex-specific architectures exist across other AD-vulnerable regions. Third, snRNA-seq cannot differentiate true changes in cell abundance from transcriptional shifts affecting cluster identity. Fourth, endogenous estrogen levels were not available, precluding direct ligand-receptor correlation analyses. Finally, although relatively large for human single-cell datasets, the cohort size warrants validation in larger independent studies.\u003c/p\u003e\n\u003cp\u003eFuture work should functionally test whether restoring estrogen pathway coordination or astrocytic APOE expression can ameliorate female-specific metabolic and synaptic vulnerabilities. Longitudinal single-cell and spatial transcriptomic studies are needed to determine when these sex-divergent trajectories first emerge. More broadly, the distinct molecular signatures identified here—homeostatic failure in females and immune activation in males—should be evaluated as potential biomarkers for sex-informed patient stratification and as targets for personalized therapy strategies.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, this study provides a comprehensive sex-stratified single-nucleus transcriptomic atlas of the human prefrontal cortex in Alzheimer\u0026apos;s disease. Our findings demonstrate that biological sex fundamentally shapes the cellular and molecular landscape of AD. We identify two distinct pathological trajectories. In females, the disease is characterized by a widespread homeostatic collapse across metabolic, transcriptional, and synaptic domains. In males, it is dominated by sustained neuroimmune activation, while core metabolic functions remain relatively preserved. We further uncover a functional uncoupling of estrogen signaling in the female AD brain. Additionally, we link female-specific astrocytic \u003cem\u003eAPOE\u003c/em\u003e downregulation to the amplified genetic risk conferred by the \u003cem\u003eAPOE \u0026epsilon;4\u003c/em\u003e allele. Collectively, these findings position biological sex as a primary determinant of cellular vulnerability in AD. They strongly advocate for integrating sex-stratified frameworks into the development of precision diagnostics and targeted therapies.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eData acquisition and cohort curation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSingle-nucleus RNA sequencing (snRNA-seq) data from human prefrontal cortex were obtained from the integrated, pre-processed dataset published by Li et al. [13], which harmonized three foundational AD cohorts [14-16]. Age metadata were manually retrieved from the original publications and appended to the integrated dataset. The curated cohort included 75 individuals (37 AD, 38 control) with complete annotation for age, sex, disease status, neuropathology (Braak, CERAD), and APOE genotype, encompassing a total of 179,690 high-quality nuclei (Fig. 1). All analyses in this study were conducted exclusively on these harmonized human datasets.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePre-processing, quality control, and cell-type annotation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe used the pre-integrated Seurat object from Li et al. without repeating batch correction to preserve consistency with the published harmonization workflow. As to maximize sensitivity for differential expression (DE) analysis, genes expressed in fewer than 3 nuclei across the dataset were filtered out. Major brain cell types were defined according to the original publication and validated using canonical marker genes: astrocytes (Ast), excitatory neurons (ex.neu; marker \u003cem\u003eSLC17A7\u003c/em\u003e) [34, 35], inhibitory neurons (in.neu), microglia (Mic; marker \u003cem\u003eCSF1R\u003c/em\u003e) [36], oligodendrocytes (Oli; marker \u003cem\u003ePLP1\u003c/em\u003e) [37], and oligodendrocyte precursor cells (OPC). All annotations were confirmed via marker-based heatmaps and cluster inspection (Fig. 2h).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDifferential expression analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSex-stratified DE analysis were conducted within each cell type using Seurat\u0026rsquo;s \u003cem\u003eFindMarkers\u003c/em\u003e function with default setting. Comparisons were made between AD and CTL samples within each sex, yielding male-specific and female-specific DEGs. Differentially expressed genes (DEGs) were defined by an absolute log₂ fold-change (avg_log2FC) \u0026gt; 0.58 (\u0026asymp;1.5-fold) and an adjusted p-value (p_val_adj) \u0026lt; 0.05 (Benjamini-Hochberg correction). DEGs were visualized using volcano plots and summarized by Venn diagrams to quantify sex specificity (Fig. 3).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunctional enrichment analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGene Ontology (GO) and KEGG pathway enrichment analyses were performed using \u003cem\u003eclusterProfiler\u003c/em\u003e with default background gene sets. Sex-specific DEG lists were analyzed separately for each cell type. Sex-specific DEGs lists were analyzed separately for each cell type. Gene Set Enrichment Analysis (GSEA) was performed for pathways correlated with \u003cem\u003eAPOE\u003c/em\u003e expression in female AD astrocytes. A ranked gene list was generated based on Spearman correlation coefficients with \u003cem\u003eAPOE\u003c/em\u003e, and enrichment was tested against the HALLMARK gene sets from the\u0026nbsp;msigdbr\u0026nbsp;package using default parameters (Fig. 6c, d).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFocused molecular and cellular analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eKey gene expression profiling\u003c/em\u003e\u003c/strong\u003e\u003cem\u003e:\u0026nbsp;\u003c/em\u003eA curated panel of AD-relevant genes (e.g., \u003cem\u003eTREM2\u003c/em\u003e, \u003cem\u003eTYROBP\u003c/em\u003e, \u003cem\u003eAPOE\u003c/em\u003e, \u003cem\u003eC1QA\u003c/em\u003e, \u003cem\u003eIL1B\u003c/em\u003e) was selected based on prior human genetics and functional studies [38-40]. Sex-stratified log2 fold-change values were quantified across all major cell types (Fig. 4).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eMicroglial functional module scoring:\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eThree functional domains were defined using curated gene sets: i) \u003cstrong\u003emetabolic activity\u003c/strong\u003e (oxidative phosphorylation genes from KEGG pathway hsa00190) [41], ii) \u003cstrong\u003eclearance capacity\u003c/strong\u003e (phagocytic receptors, e.g., \u003cem\u003eTREM2\u003c/em\u003e, \u003cem\u003eTYROBP\u003c/em\u003e) [42, 43], and iii) \u003cstrong\u003eimmune activation\u003c/strong\u003e (inflammatory cytokines and inflammasome components, e.g., \u003cem\u003eIL1B\u003c/em\u003e, \u003cem\u003eNLRP3\u003c/em\u003e) [42, 44]. Module scores were calculated using Seurat\u0026rsquo;s AddModuleScore, representing the average normalized expression across genes within each pathway (Fig. 7a).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eEstrogen signaling analysis:\u0026nbsp;\u003c/em\u003e\u003c/strong\u003eA core estrogen signaling gene set was curated to capture canonical receptor signaling and downstream neuroprotective programs, including: receptors (\u003cem\u003eESR1\u003c/em\u003e, \u003cem\u003eESR2\u003c/em\u003e, \u003cem\u003ePGR\u003c/em\u003e), primary transcriptional target (\u003cem\u003eGREB1\u003c/em\u003e), and downstream effectors involved in neuroprotection (\u003cem\u003eBDNF\u003c/em\u003e, \u003cem\u003eSIRT1\u003c/em\u003e, \u003cem\u003eBCL2\u003c/em\u003e), cell proliferation (\u003cem\u003eCCND1\u003c/em\u003e, \u003cem\u003eMYC\u003c/em\u003e), and neuronal function (\u003cem\u003eNRCAM\u003c/em\u003e, \u003cem\u003eVEGFA\u003c/em\u003e) [45-47]. Pairwise Spearman correlations among these genes were computed. Gene Set Variation Analysis (GSVA) was employed to calculate enrichment scores for early and late estrogen response pathways across cell types (Fig. 5).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eSynaptic gene analysis in excitatory neurons:\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eSynaptic gene sets were defined based on established synaptic biology literature [48-51] for:\u003c/p\u003e\n\u003cp\u003ei) \u003cstrong\u003epostsynaptic density\u003c/strong\u003e (e.g., \u003cem\u003eDLG1-4,\u0026nbsp;GRIN1/2A/2B,\u0026nbsp;SHANK1-3\u003c/em\u003e),\u003c/p\u003e\n\u003cp\u003eii) \u003cstrong\u003epresynaptic active zone\u003c/strong\u003e (e.g., \u003cem\u003eSNAP25\u003c/em\u003e, \u003cem\u003eSYT1\u003c/em\u003e, \u003cem\u003eVAMP2\u003c/em\u003e, \u003cem\u003eNRXN1-3\u003c/em\u003e), and\u003c/p\u003e\n\u003cp\u003eiii) \u003cstrong\u003esynaptic vesicle\u003c/strong\u003e components (e.g.,\u003cem\u003e\u0026nbsp;SV2A-C, SYT1-5\u003c/em\u003e). Eight synaptic DEGs from female excitatory neurons were identified and combined with four biologically important synaptic genes, yielding 12 focal genes for sex-specific synaptic profiling (Fig. 8).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eCell-cell communication analysis:\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eIntercellular signaling networks were inferred using a custom workflow based on the\u0026nbsp;CellChatDB.human\u0026nbsp;ligand-receptor database. For each experimental group (F_AD, F_CTL, M_AD, M_CTL), the average expression of each ligand and receptor was calculated per cell type. The communication strength from cell type A to B was calculated as the sum of the products\u0026nbsp;(average ligand expression in A) \u0026times; (average receptor expression in B)\u0026nbsp;across all relevant ligand-receptor pairs in database. This method generates probabilistic communication networks rather than validated physical interactions. Network-level comparisons and differential signaling matrices were constructed as shown in Fig. 9.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDifferences in cell type proportions were assessed using two-sample z-tests for proportions, with Benjamini-Hochberg (BH) correction. Group comparisons of continuous variables (e.g., functional scores, gene expression) were performed using two-sided Wilcoxon rank-sum tests. For correlation analyses, Spearman\u0026apos;s rank correlation was used for gene-gene correlations and Pearson\u0026apos;s correlation for functional module associations. Multiple testing correction was applied where appropriate using the BH method. All statistical analyses were performed in R (v4.3.0).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSoftware\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAnalysis was conducted in R (v4.3.0). Primary analysis and related visualizations mainly utilized the following packages:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSeurat\u0026nbsp;(v5.0.0)\u003c/em\u003e for single-nucleus data handling;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cem\u003eclusterProfiler\u003c/em\u003e and \u003cem\u003eenrichplot\u003c/em\u003e for functional enrichment analysis; GSVA for gene set variation analysis; \u003cem\u003emsigdbr\u003c/em\u003e for curated gene set collections; \u003cem\u003eggplot2\u003c/em\u003e, \u003cem\u003eggpubr\u003c/em\u003e, and \u003cem\u003eComplexHeatmap\u003c/em\u003e for figure generation. A comprehensive list of dependencies is documented within the associated analysis code.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eAlzheimer\u0026rsquo;s disease (AD)\u003c/p\u003e\n\u003cp\u003eApolipoprotein E \u0026epsilon;4 (APOE \u0026epsilon;4)\u003c/p\u003e\n\u003cp\u003eHormone replacement therapy (HRT)\u003c/p\u003e\n\u003cp\u003eWomen\u0026rsquo;s Health Initiative Memory Study (WHIMS)\u003c/p\u003e\n\u003cp\u003eMild cognitive impairment (MCI)\u003c/p\u003e\n\u003cp\u003eEstrogen receptor alpha (ESR1)\u003c/p\u003e\n\u003cp\u003eSIngle-nucleus RNA sequencing (snRNA-seq)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eExcitatory neurons (ExN)\u003c/p\u003e\n\u003cp\u003eInhibitory neurons (InN)\u003c/p\u003e\n\u003cp\u003eAstrocytes (Ast)\u003c/p\u003e\n\u003cp\u003eMicroglia (Mic)\u003c/p\u003e\n\u003cp\u003eOligodendrocytes (Oli)\u003c/p\u003e\n\u003cp\u003eOligodendrocyte precursor cells (OPC)\u003c/p\u003e\n\u003cp\u003eUniform manifold approximation and projection (UMAP)\u003c/p\u003e\n\u003cp\u003eDifferentially expressed genes (DEGs)\u003c/p\u003e\n\u003cp\u003eUniform Manifold Approximation and Projection (UMAP)\u003c/p\u003e\n\u003cp\u003eGene Ontology (GO)\u003c/p\u003e\n\u003cp\u003eGene Set Variation Analysis (GSVA)\u003c/p\u003e\n\u003cp\u003eGene set enrichment analysis (GSEA)\u003c/p\u003e\n\u003cp\u003eHormone replacement therapy (HRT)\u003c/p\u003e\n\u003cp\u003eDifferential expression (DE)\u003c/p\u003e\n\u003cp\u003eBenjamini-Hochberg (BH)\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable. This study reanalyzed previously published, de-identified human postmortem brain snRNA-seq datasets. Ethical approvals and consent acquisition procedures were obtained by the original studies as reported in their respective publications. No new human or animal samples were collected for the present work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCustom R scripts used for analysis and figure generation are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe snRNA-seq dataset analyzed in this study was obtained from the integrated, preprocessed resource published by Li et al. (2025). All underlying cohort data are available from the original publications and associated repositories. Accession identifiers and download instructions are provided in the Li et al. study and the source cohort papers.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research received no external funding.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eZiyi Zhou was responsible for the research design, methodology development, formal analysis, data collection and management, as well as drafting the initial manuscript. Chenxi Jiang contributed to the organization of literature review, refined analysis methods, and revised the manuscript. Zhenzhen Chen contributed to the manuscript\u0026rsquo;s revision, provided oversight and guidance throughout the research process, and managed the overall project. All authors have reviewed and approved the final version for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eLivingston G, Huntley J, Sommerlad A, Ames D, Ballard C, Banerjee S, et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. The lancet. 2020;396(10248):413-46.\u003c/li\u003e\n\u003cli\u003eGauthier S, Albert M, Fox N, Goedert M, Kivipelto M, Mestre-Ferrandiz J, et al. Why has therapy development for dementia failed in the last two decades? Alzheimer\u0026apos;s \u0026amp; Dementia. 2016;12(1):60-4.\u003c/li\u003e\n\u003cli\u003eFerretti MT, Iulita MF, Cavedo E, Chiesa PA, Schumacher Dimech A, Santuccione Chadha A, et al. Sex differences in Alzheimer disease\u0026mdash;the gateway to precision medicine. Nature Reviews Neurology. 2018;14(8):457-69.\u003c/li\u003e\n\u003cli\u003eReitz C, Brayne C, Mayeux R. Epidemiology of Alzheimer disease. Nature Reviews Neurology. 2011;7(3):137-52.\u003c/li\u003e\n\u003cli\u003eLin KA, Choudhury KR, Rathakrishnan BG, Marks DM, Petrella JR, Doraiswamy PM, et al. Marked gender differences in progression of mild cognitive impairment over 8 years. Alzheimer\u0026apos;s \u0026amp; dementia: translational research \u0026amp; clinical interventions. 2015;1(2):103-10.\u003c/li\u003e\n\u003cli\u003eNeu SC, Pa J, Kukull W, Beekly D, Kuzma A, Gangadharan P, et al. Apolipoprotein E genotype and sex risk factors for Alzheimer disease: a meta-analysis. JAMA neurology. 2017;74(10):1178-89.\u003c/li\u003e\n\u003cli\u003eShumaker SA, Legault C, Rapp SR, Thal L, Wallace RB, Ockene JK, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women\u0026apos;s Health Initiative Memory Study: a randomized controlled trial. Jama. 2003;289(20):2651-62.\u003c/li\u003e\n\u003cli\u003eShumaker SA, Legault C, Kuller L, Rapp SR, Thal L, Lane DS, et al. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women\u0026apos;s Health Initiative Memory Study. Jama. 2004;291(24):2947-58.\u003c/li\u003e\n\u003cli\u003eYanguas‐Cas\u0026aacute;s N, Crespo‐Castrillo A, Arevalo MA, Garcia‐Segura LM. Aging and sex: Impact on microglia phagocytosis. Aging Cell. 2020;19(8):e13182.\u003c/li\u003e\n\u003cli\u003eMosconi L, Berti V, Dyke J, Schelbaum E, Jett S, Loughlin L, et al. Menopause impacts human brain structure, connectivity, energy metabolism, and amyloid-beta deposition. Scientific reports. 2021;11(1):10867.\u003c/li\u003e\n\u003cli\u003eJack Jr CR, Wiste HJ, Weigand SD, Knopman DS, Vemuri P, Mielke MM, et al. Age, sex, and APOE \u0026epsilon;4 effects on memory, brain structure, and \u0026beta;-amyloid across the adult life span. JAMA neurology. 2015;72(5):511-9.\u003c/li\u003e\n\u003cli\u003eBuckley RF, Mormino EC, Rabin JS, Hohman TJ, Landau S, Hanseeuw BJ, et al. Sex differences in the association of global amyloid and regional tau deposition measured by positron emission tomography in clinically normal older adults. JAMA neurology. 2019;76(5):542-51.\u003c/li\u003e\n\u003cli\u003eLi Y, Serras CP, Blumenfeld J, Xie M, Hao Y, Deng E, et al. Cell-type-directed network-correcting combination therapy for Alzheimer\u0026rsquo;s disease. Cell. 2025;188(20):5516-34. e18.\u003c/li\u003e\n\u003cli\u003eMathys H, Davila-Velderrain J, Peng Z, Gao F, Mohammadi S, Young JZ, et al. Single-cell transcriptomic analysis of Alzheimer\u0026rsquo;s disease. Nature. 2019;570(7761):332-7.\u003c/li\u003e\n\u003cli\u003eZhou Y, Song WM, Andhey PS, Swain A, Levy T, Miller KR, et al. Human and mouse single-nucleus transcriptomics reveal TREM2-dependent and TREM2-independent cellular responses in Alzheimer\u0026rsquo;s disease. Nature medicine. 2020;26(1):131-42.\u003c/li\u003e\n\u003cli\u003eLau S-F, Cao H, Fu AK, Ip NY. Single-nucleus transcriptome analysis reveals dysregulation of angiogenic endothelial cells and neuroprotective glia in Alzheimer\u0026rsquo;s disease. Proceedings of the National Academy of Sciences. 2020;117(41):25800-9.\u003c/li\u003e\n\u003cli\u003ePark J-C, Lim H, Byun MS, Yi D, Byeon G, Jung G, et al. Sex differences in the progression of glucose metabolism dysfunction in Alzheimer\u0026rsquo;s disease. Experimental \u0026amp; Molecular Medicine. 2023;55(5):1023-32.\u003c/li\u003e\n\u003cli\u003eNerattini M, Abenavoli EM, Caramelli F, Giacomucci G, Nacmias B, Mossello E, et al. Sex differences in brain glucose metabolism in alzheimer\u0026rsquo;s disease: A voxel-based study. GeroScience. 2025:1-11.\u003c/li\u003e\n\u003cli\u003eScheyer O, Rahman A, Hristov H, Berkowitz C, Isaacson R, Diaz Brinton R, et al. Female sex and Alzheimer\u0026rsquo;s risk: the menopause connection. The journal of prevention of Alzheimer\u0026apos;s disease. 2018;5(4):225-30.\u003c/li\u003e\n\u003cli\u003eLopez-Lee C, Torres ERS, Carling G, Gan L. Mechanisms of sex differences in Alzheimer\u0026rsquo;s disease. Neuron. 2024;112(8):1208-21.\u003c/li\u003e\n\u003cli\u003eSochocka M, Ochnik M, Sobczyński M, Orzechowska B, Leszek J. Sex differences in innate immune response of peripheral blood leukocytes of Alzheimer\u0026rsquo;s disease patients. Archivum Immunologiae et Therapiae Experimentalis. 2022;70(1):16.\u003c/li\u003e\n\u003cli\u003eJanelidze S, Mattsson N, Stomrud E, Lindberg O, Palmqvist S, Zetterberg H, et al. CSF biomarkers of neuroinflammation and cerebrovascular dysfunction in early Alzheimer disease. Neurology. 2018;91(9):e867-e77.\u003c/li\u003e\n\u003cli\u003eDuarte-Guterman P, Albert AY, Inkster AM, Barha CK, Galea LA, Initiative AsDN. Inflammation in Alzheimer\u0026rsquo;s disease: do sex and APOE matter? Journal of Alzheimer\u0026rsquo;s Disease. 2020;78(2):627-41.\u003c/li\u003e\n\u003cli\u003eBiechele G, Rauchmann B-S, Janowitz D, Buerger K, Franzmeier N, Weidinger E, et al. Associations between sex, body mass index and the individual microglial response in Alzheimer\u0026rsquo;s disease. Journal of neuroinflammation. 2024;21(1):30.\u003c/li\u003e\n\u003cli\u003ePerneczky R, Drzezga A, Diehl-Schmid J, Li Y, Kurz A. Gender differences in brain reserve: an 18F-FDG PET study in Alzheimer\u0026apos;s disease. Journal of neurology. 2007;254(10):1395-400.\u003c/li\u003e\n\u003cli\u003eIshunina TA, Fischer DF, Swaab DF. Estrogen receptor \u0026alpha; and its splice variants in the hippocampus in aging and Alzheimer\u0026apos;s disease. Neurobiology of aging. 2007;28(11):1670-81.\u003c/li\u003e\n\u003cli\u003eFoster TC. Role of estrogen receptor alpha and beta expression and signaling on cognitive function during aging. Hippocampus. 2012;22(4):656-69.\u003c/li\u003e\n\u003cli\u003eSun LL, Yang SL, Sun H, Li WD, Duan SR. Molecular differences in Alzheimer\u0026apos;s disease between male and female patients determined by integrative network analysis. Journal of Cellular and Molecular Medicine. 2019;23(1):47-58.\u003c/li\u003e\n\u003cli\u003eSantiago JA, Quinn JP, Potashkin JA. Sex-specific transcriptional rewiring in the brain of Alzheimer\u0026rsquo;s disease patients. Frontiers in Aging Neuroscience. 2022;14:1009368.\u003c/li\u003e\n\u003cli\u003eSampedro F, Vilaplana E, De Leon MJ, Alcolea D, Pegueroles J, Montal V, et al. APOE-by-sex interactions on brain structure and metabolism in healthy elderly controls. Oncotarget. 2015;6(29):26663.\u003c/li\u003e\n\u003cli\u003eHohman TJ, Dumitrescu L, Barnes LL, Thambisetty M, Beecham G, Kunkle B, et al. Sex-specific association of apolipoprotein E with cerebrospinal fluid levels of tau. JAMA neurology. 2018;75(8):989-98.\u003c/li\u003e\n\u003cli\u003eMhatre-Winters I, Eid A, Han Y, Tieu K, Richardson JR. Sex and APOE genotype alter the basal and induced inflammatory states of primary astrocytes from humanized targeted replacement mice. ASN neuro. 2023;15(1):17590914221144549.\u003c/li\u003e\n\u003cli\u003eStephen T, Breningstall B, Suresh S, McGill C, Pike C. APOE genotype and biological sex regulate astroglial interactions with amyloid plaques in Alzheimer\u0026rsquo;s disease mice. Journal of neuroinflammation. 2022;19(1):286.\u003c/li\u003e\n\u003cli\u003eYao Z, Liu H, Xie F, Fischer S, Adkins RS, Aldridge AI, et al. A transcriptomic and epigenomic cell atlas of the mouse primary motor cortex. Nature. 2021;598(7879):103-10.\u003c/li\u003e\n\u003cli\u003eFremeau Jr RT, Kam K, Qureshi T, Johnson J, Copenhagen DR, Storm-Mathisen J, et al. Vesicular glutamate transporters 1 and 2 target to functionally distinct synaptic release sites. Science. 2004;304(5678):1815-9.\u003c/li\u003e\n\u003cli\u003eGinhoux F, Greter M, Leboeuf M, Nandi S, See P, Gokhan S, et al. Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science. 2010;330(6005):841-5.\u003c/li\u003e\n\u003cli\u003eKlugmann M, Schwab MH, P\u0026uuml;hlhofer A, Schneider A, Zimmermann F, Griffiths IR, et al. Assembly of CNS myelin in the absence of proteolipid protein. Neuron. 1997;18(1):59-70.\u003c/li\u003e\n\u003cli\u003eLue L-F, Schmitz C, Walker D. What happens to microglial TREM2 in Alzheimer\u0026rsquo;s disease: Immunoregulatory turned into immunopathogenic? Neuroscience. 2015;302:138-50.\u003c/li\u003e\n\u003cli\u003eHaure-Mirande J-V, Audrain M, Ehrlich ME, Gandy S. Microglial TYROBP/DAP12 in Alzheimer\u0026rsquo;s disease: Transduction of physiological and pathological signals across TREM2. Molecular neurodegeneration. 2022;17(1):55.\u003c/li\u003e\n\u003cli\u003eVillegas-Llerena C, Phillips A, Garcia-Reitboeck P, Hardy J, Pocock JM. Microglial genes regulating neuroinflammation in the progression of Alzheimer\u0026apos;s disease. Current opinion in neurobiology. 2016;36:74-81.\u003c/li\u003e\n\u003cli\u003eKanehisa M, Furumichi M, Sato Y, Kawashima M, Ishiguro-Watanabe M. KEGG for taxonomy-based analysis of pathways and genomes. Nucleic acids research. 2023;51(D1):D587-D92.\u003c/li\u003e\n\u003cli\u003eKeren-Shaul H, Spinrad A, Weiner A, Matcovitch-Natan O, Dvir-Szternfeld R, Ulland TK, et al. A unique microglia type associated with restricting development of Alzheimer\u0026rsquo;s disease. Cell. 2017;169(7):1276-90. e17.\u003c/li\u003e\n\u003cli\u003eUlland TK, Song WM, Huang SC-C, Ulrich JD, Sergushichev A, Beatty WL, et al. TREM2 maintains microglial metabolic fitness in Alzheimer\u0026rsquo;s disease. Cell. 2017;170(4):649-63. e13.\u003c/li\u003e\n\u003cli\u003eFriedman BA, Srinivasan K, Ayalon G, Meilandt WJ, Lin H, Huntley MA, et al. Diverse brain myeloid expression profiles reveal distinct microglial activation states and aspects of Alzheimer\u0026rsquo;s disease not evident in mouse models. Cell reports. 2018;22(3):832-47.\u003c/li\u003e\n\u003cli\u003eToumba M, Kythreotis A, Panayiotou K, Skordis N. Estrogen receptor signaling and targets: Bones, breasts and brain. Molecular Medicine Reports. 2024;30(2):144.\u003c/li\u003e\n\u003cli\u003eMarquardt RM, Kim TH, Shin J-H, Jeong J-W. Progesterone and estrogen signaling in the endometrium: what goes wrong in endometriosis? International journal of molecular sciences. 2019;20(15):3822.\u003c/li\u003e\n\u003cli\u003ePark H, Pagan L, Tan O, Fadiel A, Demir N, Huang K, et al. Estradiol regulates expression of polysialated neural cell adhesion molecule by human vascular endothelial cells. Reproductive Sciences. 2010;17(12):1090-8.\u003c/li\u003e\n\u003cli\u003eS\u0026uuml;dhof TC. Neurotransmitter release: the last millisecond in the life of a synaptic vesicle. Neuron. 2013;80(3):675-90.\u003c/li\u003e\n\u003cli\u003eFatemi SH, Eschenlauer A, Aman J, Folsom TD, Chekouo T. Quantitative proteomics of dorsolateral prefrontal cortex reveals an early pattern of synaptic dysmaturation in children with idiopathic autism. Cerebral Cortex. 2024;34(13):161-71.\u003c/li\u003e\n\u003cli\u003eGomez AM, Traunm\u0026uuml;ller L, Scheiffele P. Neurexins: molecular codes for shaping neuronal synapses. Nature Reviews Neuroscience. 2021;22(3):137-51.\u003c/li\u003e\n\u003cli\u003eBonnycastle K, Davenport EC, Cousin MA. Presynaptic dysfunction in neurodevelopmental disorders: Insights from the synaptic vesicle life cycle. Journal of Neurochemistry. 2021;157(2):179-207.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Alzheimer’s disease, snRNA-seq, Sex differences, Homeostatic failure, Neuroimmune activation, Microglial metabolism, Astrocytic APOE, Estrogen signaling decoupling","lastPublishedDoi":"10.21203/rs.3.rs-9504119/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9504119/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eAlzheimer\u0026rsquo;s disease (AD) disproportionately affects females in both prevalence and clinical progression, yet the underlying sex-dependent cellular programs remain poorly understood. This study aims to delineate the sex-specific pathological landscape of AD at a single-nucleus resolution to identify the biological drivers of these disparities. We analyzed a large-scale integrated snRNA-seq dataset from the human prefrontal cortex, encompassing 179,690 nuclei harmonized across three foundational AD cohorts.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eOur analysis revealed a sex-dichotomous pathological landscape. In females, AD is characterized by a systemic erosion of homeostatic programs: microglial oxidative phosphorylation drops sharply, astrocytic \u003cem\u003eAPOE\u003c/em\u003e expression is significantly downregulated, and the presynaptic compartment is severely depleted. In contrast, males exhibit a distinct pathological axis dominated by sustained neuroimmune activation while metabolic and synaptic capacities remain relatively preserved. Furthermore, we identified a functional collapse of the estrogen signaling axis in females; despite a paradoxical upregulation of \u003cem\u003eESR1\u003c/em\u003e, its downstream neuroprotective targets remain inactive. The female-specific downregulation of astrocytic \u003cem\u003eAPOE\u003c/em\u003e potentially explains the amplified genetic risk associated with the \u003cem\u003eAPOE ε4\u003c/em\u003e allele in women.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eOur findings demonstrate that biological sex is a primary determinant of cellular vulnerability in AD. The identified transcriptional uncoupling of estrogen signaling and sex-specific metabolic-immune failure underscore the necessity of sex-stratified approaches in AD precision medicine and the design of targeted interventions.\u003c/p\u003e","manuscriptTitle":"Single-Nucleus Transcriptomics Analysis Identifies Sex-Dichotomous Pathological Axes in Alzheimer’s Disease: Female Homeostatic Failure versus Male Neuroimmune Activation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-13 10:59:43","doi":"10.21203/rs.3.rs-9504119/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"42adfe2a-8b57-47da-bb1b-e94c30d990a1","owner":[],"postedDate":"May 13th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Rejected","date":"2026-05-15T10:59:17+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-05-15T11:10:20+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-13 10:59:43","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9504119","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9504119","identity":"rs-9504119","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}