Brain wide engraftment of human monocytes induces neurodegeneration and cognitive dysfunction

Brain wide engraftment of human monocytes induces neurodegeneration and cognitive dysfunction | 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 Biological Sciences - Article Brain wide engraftment of human monocytes induces neurodegeneration and cognitive dysfunction Mathew Blurton-Jones, Hayk Davtyan, Jean Paul Chadarevian, Anita Lakatos, and 20 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7762816/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Transplantation of peripherally derived myeloid cells in combination with microglia depletion is increasingly being examined as a potential therapy for various neurological disorders.1-9 However, primitive yolk sac derived microglia and definitive bone marrow derived monocytes and macrophages exhibit distinct developmental ontogenies.10-12 Important differences between murine and human myeloid cells further obscure the translational potential of this approach.8,13-16 To understand the impact of peripherally derived human myeloid cells on brain function and health we utilized a xenotolerant mouse model that lacks endogenous microglia. Using a combination of spatial sequencing, proteomics, histological, biochemical, and machine learning behavioral approaches we then examined the impact of brain wide engraftment of human induced pluripotent stem cell-derived microglia, which mimic the primitive ontogeny of microglia, in comparison to definitive bone marrow derived monocytes. Here we show that engraftment of human monocytes, but not microglia, promotes a chronic proinflammatory state associated with astrogliosis, demyelination, and synaptic and neuronal loss. These robust changes in myeloid cell activation state and brain homeostasis are in turn accompanied by significant impairments in cognitive function. Together with the accompanying study by Wang and colleagues, we demonstrate the critical impact of myeloid cell ontogeny on neuronal health and function, providing important implications for the rapidly progressing development and translation of brain-wide microglial replacement therapies.1-5 Biological sciences/Neuroscience/Neuroimmunology Biological sciences/Immunology/Innate immune cells/Monocytes and macrophages/Microglial cells Biological sciences/Stem cells/Haematopoietic stem cells ontogeny microglia monocyte myeloid macrophage iPSC pluripotent xenotransplantation hematopoietic stem and progenitor cells (HSPC) bone marrow transplantation (BMT) Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Main Recent studies have examined the therapeutic potential of bone marrow transplantation in mouse models of Alzheimer’s disease, frontotemporal dementia, and lysosomal storage diseases. 1-5 By combining myeloablative preconditioning with microglial-depleting CSF1R antagonists and hematopoietic stem and progenitor cell (HSPC) transplantation, these studies have shown that brain wide replacement of microglia with murine bone marrow derived ‘microglia-like cells (MLCs) leads to promising reductions in disease-associated pathologies. However, prior studies have also clearly demonstrated that despite long-term brain engraftment, peripherally derived murine MLCs remain transcriptionally, morphologically, and functionally distinct from microglia. 6-9 Given these results along with the well-established differences in microglial versus peripheral myeloid cell ontogeny, 10-12 it is important to understand the long-term consequences of replacing microglia with MLCs. As human and mouse microglia also exhibit significant differences, especially in regard to neurodegeneration-associated genes, 8,13-16 it is equally important to investigate the effects of brain wide human MLC engraftment in appropriate model systems before this approach is translated into clinical practice. Here we utilized xenotolerant FIRE mice, a mouse model that enables complete humanization of the central nervous system (CNS) microglial compartment, to examine the long-term consequences of replacing endogenous microglia with definitive HSPC-derived human monocytes or induced pluripotent stem cell (iPSC)-derived microglia (iMG) that are generated via a yolk sac-like primitive hematopoietic progenitor cell (HPC) intermediate (Fig. 1a). 17 Fourteen weeks later, brains were examined via transcriptomic, proteomic, biochemical, and histological approaches to determine the impact of myeloid cell ontogeny on brain health and function. Together with the accompanying study by Wang et. al., we demonstrate robust and reproducible ontogeny-dependent differences in the activation state of definitive HSPC-derived monocytes versus primitive HPC derived microglia. Importantly, these ontogeny-dependent differences mimic those observed in a human neurodegenerative disease and are further accompanied by impairments in neuronal and glial homeostasis, culminating in pronounced cognitive dysfunction. Collectively, our findings suggest that transplantation of iPSC-microglia that closely mimic the primitive developmental ontogeny of human microglia could offer a safer alternative to definitive HSPC transplantation for the treatment of neurological diseases. Human monocyte engraftment induces a proinflammatory state We previously developed xenotolerant FIRE mice that enable long-term brain wide engraftment of human microglia derived from pluripotent stem cells or surgical resection and demonstrated that engrafted iPSC-derived microglia are highly similar to brain derived human microglia. 18,19 At 2 months of age, male mice were transplanted with human blood monocytes or iMG derived from multiple male control subjects. Fourteen weeks later, brains were examined using a combination of transcriptomic, proteomic, biochemical, and histological approaches (Fig. 1b). Both microglia and monocytes migrated well beyond the hippocampal injection site, leading to similar levels of brain wide engraftment (Fig. 1c-f). Consistent with our prior studies, 15,19 engrafted human iMG adopted a highly ramified cell morphology consistent with the adoption of a homeostatic state (Fig. 1g). While monocytes also exhibited morphological changes with engraftment, branching complexity remained significantly lower than that of microglia and was frequently accompanied by a more elongated morphology (Fig. 1h-j). To examine the impact of the brain environment on microglial and monocyte gene expression we utilized Visium spatial sequencing. Clustering across all six samples (n=3/group) revealed 13 regionally distinct signatures that corresponded to well established anatomical regions including upper and lower cortical layers, hippocampal subfields, the amygdala, hypothalamus, and white matter tracts (Fig. 2a,b). To maximize rigor, we focused our analysis on the intersection of differentially expressed genes (DEGs) detected using both a spot-based and pseudobulk analyses of the dataset (Fig. 2c). The use of xenotolerant FIRE mice that lack microglia coupled with a human targeted Visium probe set provided additional confidence that the great majority of DEGs detected in this experiment were derived from engrafted human myeloid cells. Consistent with this expectation, pseudobulk comparison between microglia and monocyte-engrafted tissue slices detected 1768 DEGs (LogFc>1, FDR<0.05), many of which represented well-established myeloid genes (Fig. 2d, Supplementary Table 1). While microglial engrafted tissue exhibited high levels of canonical homeostatic microglial genes ( P2RY12, SALL1, CX3CR1, VSIR ), monocyte-engrafted mice instead exhibited greatly increased expression of genes associated with proinflammatory activation of macrophages such as G PNMB, CD9, APOE, LGALS3, STAT1, CD163, and CTSD (Fig. 2d). Spatial maps of P2RY12, APOE, and CD74 demonstrate the broadly increased expression of P2RY12 in microglial samples in contrast to the greatly increased expression of APOE and CD74 in monocyte engrafted brains (Fig. 2e). Spatial mapping of representative genes across all six samples further demonstrated the consistent enrichment of homeostatic microglial genes ( P2RY12, SALL1 ) versus monocyte/macrophage associated genes ( CD14, APOE, CD74, MX1 ) across all replicates (Extended Data Fig. 1a). Violin plots of representative DEGs further demonstrate the enrichment of homeostatic genes in microglia and pro-inflammatory genes in monocytes (Fig. 2f, Supplementary Table 1). Brain engrafted monocytes exhibit a DAM-like signature We were surprised to find that many of the genes enriched in monocyte engrafted mice (Fig. 2f) corresponded closely with the signature of ‘disease associated microglia (DAM)’, a transcriptional state that has consistently been observed in models of Alzheimer’s Disease (AD). 20 In previous studies, we and others have similarly reported the induction of a DAM-like signature in human xMG adjacent to beta-amyloid plaques. 15,21,22 In contrast to monocyte engrafted hFIRE mice, xMG engrafted mice in the current study , express low levels of DAM associated genes, and instead exhibit typical homeostatic microglial gene expression (Fig. 2d-f) and morphology (Fig. 1g,i,j). To further understand the potential similarity between the DAM signature and our spatial sequencing data, we used a previously published single cell sequencing (scSeq) dataset from xenografted AD mice 15,23 to generate a DAM module score. Mapping of this module score across each of the 13 brain regions identified in our spatial analysis revealed a significant enrichment of the DAM signature across all regions of the monocyte engrafted brains without evidence of plaque or tangle pathologies in comparison to very low module scores within microglia-engrafted brains (Fig. 2g). To determine whether our spatial sequencing analysis was consistent with the independently generated scSeq data presented in the accompanying study by Wang and colleagues, we compared the overlap in gene expression between both studies. This analysis revealed a highly significant overlap (P =1.98 x 10 -304 , Extended Data Fig. 1b) with over 40% of upregulated monocyte genes detected by Wang exhibiting a similar upregulation in monocytes within our spatial sequencing dataset. A Ratio-Ratio plot comparing Log2Fc expression differences between monocytes and microglia detected in both studies (Fig. 2h) likewise revealed a strong positive correlation (Pearson r=0.51, P < 2.2 x 10 -16 ). To provide a third comparison between these two datasets we generated microglial and monocyte module scores from the Wang et. al. dataset and mapped those scores onto the 13 brain regions examined in our spatial sequencing data. This comparison again revealed a strong overlap between the two datasets across all 13 brain regions (Extended Data Fig. 1c). In summary, the spatial sequencing analysis performed in the current study and the scSeq analysis performed by Wang and colleagues reveal remarkably consistent and reproducible results, especially given the technical differences between the two studies which included differing sequencing approaches, donor cells and sources, mouse colonies, and ages of transplant. Proteomic analysis further distinguishes inflammatory monocytes from homeostatic microglia In addition to our spatial sequencing analysis, rapidly frozen hemibrains from 8 mice per group were homogenized and examined by Tandem Mass Tag mass spectrometry (TMT-MS) (Fig. 3a). Our proteomic analysis revealed 1,406 differentially expressed proteins (DEPs; Adjusted P<0.05) between microglia and monocyte engrafted brains (Supplementary Table 2). Of those DEPs, 730 proteins were upregulated in monocyte engrafted brains whereas 676 downregulated DEPs were detected in monocyte engrafted mice (Extended Data Fig. 2a). However, when the top 70 DEPs were compared by heatmap, the great majority of DEGs were increased in monocyte brains, with only 4 out of 70 proteins exhibiting reduced expression in monocyte engrafted brains (Fig. 3b). Consistent with our spatial seq analysis, homoeostatic microglial proteins P2RY12, ACY3, and CRYBB1 exhibited increased expression in microglial engrafted brains, whereas monocyte engrafted brains exhibited greatly increased levels of proinflammatory myeloid and DAM-associated proteins including GPNMB, LGALS3 (Galectin 3), CD9, CD163, FTL (ferritin light chain), LIPA, ITGAX (CD11c) as well as Human leukocyte antigen (HLA) associated proteins involved in both MHC-I (HLA-A, HLA-B, B2M) and MHCII (HLA-DRA, HLA-DRB1, HLA-DPB1, CD74) antigen presentation (Fig. 3b). Violin plots of representative DEPs further demonstrate the consistent enrichment of myeloid activation-associated proteins in monocyte engrafted mice versus increased homeostatic proteins in microglia engrafted brains (Fig. 3c). Principle component analysis of our proteomics data demonstrated the consistency of these findings; revealing a clear separation between the 8 monocyte engrafted samples and 8 microglia engrafted samples along principal component 1 (PC1) representing 44.8% of the variance within the dataset (Fig. 3d). Next, to further confirm that the expression of myeloid-associated DEPs in half brain homogenates was derived from engrafted human cells, we performed immunohistochemistry (IHC) for five representative DEPs along with IBA-1 to label all myeloid cells and the human-specific nuclear marker KU80 (Fig. 3e-s). IMARIS-based quantitative analysis revealed significantly increased expression of LGALS3 (Galectin-3) within engrafted monocytes in comparison to microglia within both the cortex and hippocampus (Fig. 3e-g). In contrast, IHC analysis of the homeostatic marker P2RY12 demonstrated far greater expression within microglia (Fig. 3h,I,m). Examination of three additional DAM-associated proteins (GPNMB, HLA-DRB1, CD9) also revealed greatly increased expression within human monocytes (Fig. 3 i-k,n-s). To better define the functional pathways impacted by brain wide monocyte engraftment we performed Gene Ontology (GO) analysis of our proteomic dataset (Supplementary Table 3). Representative upregulated GO pathways included terms such as ‘Phagosome, antigen processing presentation, Lipid and atherosclerosis, efferocytosis, TNF signaling, JAK-STAT signaling, Ferroptosis, and cholesterol metabolism’ (Extended Data Fig. 3b). In contrast, GO pathways that were downregulated in monocyte engrafted mice highlighted terms such as Synaptic vesicle cycle, Cholinergic synapse, GABAergic synapse, Glutamatergic synapse, and Dopaminergic synapse (Extended Data Fig. 3c). Additional downregulated terms included Long-term potentiation (LTP), long-term depression (LTD), neurotrophin signaling, axon guidance, and longevity regulating pathways. As many of these downregulated pathways involve neuronal and synaptic proteins, we next sought to examine the potential impact of monocyte engraftment on non-myeloid cell types. Monocyte engraftment promotes the loss of neuronal, synaptic, and oligodendroglial proteins While myeloid associated proteins were highly enriched in the top 70 upregulated DEPs (Figure 3b), our proteomic analysis further revealed a large number of non-myeloid DEPs that were significantly decreased in monocyte engrafted brains (Fig. 4a, Supplementary Table 2). Consistent with our analysis of down-regulated GO terms (Extended Data Fig. 2c), many of these non-myeloid DEPs represented canonical neuronal and synaptic proteins such as Tau (MAPT), β3-tubulin (TUBB3), Synaptophysin (SYP), Synapsin-1 (SYN1), Synaptic Vesicle Glycoprotein 2A (SV2A), and PSD-95 (DLG4). Likewise, we observed significant reductions in several canonical oligodendrocyte and oligodendrocyte precursor cell (OPC) proteins including Myelin Basic Protein (MBP), Myelin Oligodendrocyte Glycoprotein (MOG), Myelin-Associated Glycoprotein (MAG), 2',3'-cyclic nucleotide 3'-phosphodiesterase (CNP), and proteolipid protein 1 (PLP-1) (Fig. 4a). To further confirm these key changes, we next measured levels of murine SYP, PSD-95, and MBP by ELISA. In each case we again observed a significant reduction in these proteins within monocyte engrafted brains (Fig. 4b), whereas microglia-engrafted mice showed no differences in comparison to ‘wild type’ hCSF1 immune-deficient controls. Monocytes increase astrogliosis and proinflammatory cytokines While the majority of non-myeloid proteins exhibited decreased expression in monocyte-engrafted brains, several DEPs that are typically enriched in astrocytes or endothelial cells exhibited elevated expression (Fig. 4a, Supplementary Table 2). For example, Glial fibrillary acid protein (GFAP), a commonly utilized marker of reactive astrogliosis, along with the astrocytic proteins ALDH1L1, S100β, and aquaporin-4 (AQP4), exhibited elevated expression in monocyte brains (Fig. 4a). ELISA analysis of GFAP and Complement 4 (C4), both of which have frequently been associated with astrogliosis further corroborated our proteomic findings (Fig. 4b). Next, to determine whether the observed increases in myeloid cell activation were further accompanied by changes in cytokine production, we measured ten cytokines via multiplex ELISA. This analysis revealed significantly increased levels of six proinflammatory cytokines within monocyte engrafted brains including TNFα, Il1β, IL-2, IL-8, IL-12, and INFγ (Fig. 4b), whereas the four remaining analytes (IL-4, IL-6, IL-10, IL-13) showed no significant differences (Extended Data Fig. 3a). Interestingly, IL-34, which along with CSF1 signals via CSF1R to promote microglial differentiation and survival, was likewise elevated within monocyte engrafted brains. Importantly, none of these 10 cytokines were elevated within human microglial engrafted brains in comparison to wildtype hCSF1 immune-deficient controls. To further determine whether there might be a relationship between the observed reductions in synaptic and oligodendroglial proteins and increased proinflammatory cytokines we next generated a correlation table to compare each of the altered brain analytes. This comparison revealed highly significant correlations between many of these analytes including strong inverse associations (blue boxes) between synaptophysin and several proinflammatory cytokines (Fig. 4c, blue boxes). Individual regression analysis of synaptophysin versus C4 and synaptophysin versus TNFα further revealed highly significant correlations of R 2 =0.719 and R 2 =0.577 respectively (Fig 4d, P≤0.0006). These same two proinflammatory analytes also exhibited significant inverse relationships with MBP levels (Fig. 4e, P≤0.0055). Lastly, to determine whether the observed elevations in proinflammatory cytokines can also be detected in the periphery, we performed the same multiplex ELISA on serum samples collected from each mouse at the time of sacrifice. This analysis revealed significant increases in all ten analytes suggesting that monocyte engraftment within the brain also leads to a heightened proinflammatory environment within the periphery (Extended Data Fig. 3b). Human monocytes promote hippocampal neurodegeneration Given that our proteomic analysis had detected significant decreases in neuronal proteins including Tau and β3-tubulin (Fig. 4a) and our spatial sequencing had revealed highly elevated levels of proinflammatory genes within the hippocampus (Fig. 2e, Extended Data Fig. 1), we performed IHC for the neuronal nuclear marker NeuN and examined the hippocampus. We were surprised to observe a dramatic loss of NeuN+ neurons within the upper blade of the dentate gyrus (Fig. 4f) in 5 of the 11 monocyte-engrafted brains examined. Because group distributions of NeuN deviated from normality, we utilized a non-parametric Kruskal–Wallis test followed by Dunn’s multiple comparisons, revealing a significant reduction in NeuN density in monocyte engrafted mice versus untransplanted ‘wildtype’ hCSF1 controls (Fig. 4g, Kruskal–Wallis: P=0.042, Dunn’s: P=0.049), but no significant differences between microglia engrafted mice and wildtype controls (Dunn’s: P>0.999). Interestingly, co-labeling with IBA-1 revealed densely packed ameboid-shaped monocytes in the same region of the upper blade where NeuN labeling had been lost in monocyte brains. We therefore wondered whether monocytes might be promoting the death and/or clearance of degenerating neurons within this region. Monocytes, macrophages, and microglia can rapidly phagocytose dying cells. In fact, an important physiological function of monocyte-derived macrophages within the spleen is to phagocytose and degrade aging and dysfunctional red blood cells. 24 However, phagocytosis proceeds rapidly, and antigenicity of internalized proteins is promptly lost by lysosomal degradation, making it challenging to reliably catch snapshots of cellular phagocytosis in the adult brain in vivo . Three recent reports have further demonstrated the difficulty of reliably quantifying microglial phagocytosis of cells, synapses, and myelin. 25-27 Nevertheless, we observed many examples of NeuN+ neurons within the hippocampus that were surrounded by ameboid-shaped monocytes that expressed high levels of the phagolysosome marker CD68 (Fig. 4h), suggesting that monocytes rapidly clear or perhaps even promote the loss of neurons within the hippocampus. Examples of monocytes with internalized synaptophysin (Extended Data Fig. 4a,b) or myelin basic protein (Extended Data Fig. 4c,d) were also readily observed within the hippocampus and corpus callosum respectively. While peptides can be rapidly cleared by the lysosome, a growing number of studies have demonstrated the accumulation and persistence of lipid droplets within microglia during neurodegenerative disease. 28-30 Monocyte-derived macrophages within the periphery are likewise well recognized to accumulate lipid droplets during diseases such as arteriosclerosis, Non-Alcoholic Steatohepatitis, and lysosomal storage diseases. 31-33 We therefore wondered whether brain engrafted monocytes might similarly exhibit lipid droplet accumulation. While IHC analysis of engrafted microglia revealed little to no accumulation of PLIN-2+ lipid droplets, engrafted monocytes exhibited a dramatic increase in lipid droplet density within the cortex and hippocampus (Extended Data Fig. 4e-h). Taken together, our findings suggest that peripherally derived human monocytes adopt a chronically pro-inflammatory signature that promotes the phagocytosis of neurons, synapses, and oligodendrocytes and the accumulation of lipid droplets. Importantly, the accompanying study by Wang and colleagues provides additional independent evidence that human myeloid cell ontogeny strongly influences neuronal and glial health and lipid droplet accumulation. The transcriptional signature of brain engrafted monocytes occurs in human neurodegenerative disease CSF1R-related disorder (CSF1R-RD) is a rare neurodegenerative disease associated with mutations in CSF1R, a receptor that is critical for microglia differentiation, survival, and function. 34,35 Also referred to as adult-onset leukoencephalopathy with axonal spheroids and pigmented glia (ALSP), patients affected by this disease exhibit decreased numbers of homeostatic microglia leading to demyelination, astrogliosis, and axonal spheroids which is accompanied by progressive cognitive and/or motor dysfunction. 36 Recent studies have examined the potential use of bone marrow transplantation (BMT) to treat CSF1R-RD with highly conflicting results. For example, one study of 15 patients with a median follow-up of 26 months, found that BMT was associated with a significant worsening of cognitive symptoms. 37 In contrast, another recent study reported that BMT slowed cognitive decline in five patients, although one of these five subjects exhibited Montreal Cognitive Assessment (MoCA) and mini mental state exam (MMSE) scores of 0-1 out of 30 throughout the study. Most recently, an elegant single nuclei sequencing (snSeq) study compared the brains of five ALSP subjects to twelve control subjects. 38 In addition to further demonstrating the loss of homeostatic microglia in ALSP brains, Du and colleagues detected a unique “ALSP-enriched microglia” signature that was virtually absent in controls. The authors further concluded that this unique cell cluster could either represent highly activated microglia or brain infiltrating “macrophages derived from peripheral monocytes”. 38 Interestingly, DEG analysis of this ‘ALSP microglia’ cluster highlighted many genes that are also enriched in our current monocyte transcriptional signature, including GPNMB, PLIN2, APOE, FTL, CD163, MITF, CHIT1, CR1, LGALS3, APOC1, ANXA2 among others (Fig. 2d,f; Supplementary Table 1). We therefore sought to directly compare our monocyte and microglia signatures to those observed in human control and ALSP brains by Du et. al. Using our spatial sequencing dataset we generated module scores from the top 200 genes upregulated in monocyte or microglia engrafted brains. These module scores were then mapped onto control and ALSP patient UMAPs that were reconstructed from Du et. al.’s snSeq dataset (Fig. 5a-f). Consistent with our previously published comparisons between brain-engrafted iPSC-microglia and human resection-derived ex vivo microglia, 15 our microglial module score overlapped closely with homeostatic microglia identified by Du and colleagues (Fig. 5b,e). In contrast, our monocyte module score was greatly enriched within the unique cluster of ‘ALSP-microglia’ identified by Du et. al., (Fig. 5f). To further compare these datasets, we performed the inverse analysis, calculating ‘control-microglia’ and ‘ALSP-microglia’ module scores from Du et. al., and demonstrating a strong enrichment of these modules in our microglia or monocyte engrafted brains respectively (Fig. 5g). In addition, we examined the log fold change in all significant myeloid DEGs detected by both studies. Regression analysis revealed a highly significant positive correlation (R 2 = 0.36, P<0.0001), further demonstrating the correspondence between Du and colleagues ‘ALSP microglia’ and our current human monocyte engrafted samples. Lastly, we mapped the module scores calculated in Fig 5g onto our Visium spatial images, further demonstrating a strong enrichment of the ALSP-microglia module in our monocyte engrafted brains and a strong enrichment of the control homeostatic module in our iMG engrafted brains (Fig. 5i,j). Collectively, these analyses strongly suggest that the myeloid activation signature observed in our monocyte engrafted mice is closely related to a dysfunctional myeloid signature associated with a rapidly progressing and devastating human neurodegenerative disease. Monocyte engraftment promotes cognitive dysfunction To determine the impact of the proinflammatory and hyperphagocytic profile of CNS-wide engrafted human monocytes on behavior, we quantified mouse exploratory behavior in the open field using Motion Sequencing (MoSeq), which uses machine learning to decompose movement into elemental motifs of action (e.g., runs, rears, turns) called “syllables”. 39 We quantified the expression of individual behavioral syllables expressed by transplanted mice at 5 months of age in an open-field arena over five consecutive days. We then subsequently performed three additional behavioral assays: rotarod, nesting, and marble burying (Fig 6a). In addition to revealing patterns of syllable usage, the MoSeq framework provides quantitative summary data r that revealed a striking difference in the position of mice transplanted with microglia versus monocytes as well as readily apparent difference in syllable usage (Fig. 6b). Consistent with pervasive behavioral differences between monocyte and microglia-transplanted mice, 78% of all syllables identified exhibited significant differences in usage in the open field (Fig 6c). Linear discriminant analysis (LDA) revealed a clear separation in the overall patten of syllable use between the transplanted mice that was stable over days (Fig. 6d). Visualization of the arena over 30 mins also revealed pronounced thigmotaxis within the monocyte engrafted mice as they paced along the arena walls (Fig. 6e). During the first minute of observation both monocyte and microglia mice exhibited typical murine behavior, exploring the border of the arena. However, over 30 minutes, monocyte engrafted mice continued to exhibit hyperactivity, showing no signs of habituation and leading to a significant increase in total distance traveled over 30 minutes (Fig. 6f,g). Sickness behavior in mice includes lethargy and reduced appetite and can be induced by proinflammatory stimuli such as lipopolysaccharide (LPS). 40 Similar to LPS, we found that monocyte engraftment leads to increased levels of multiple proinflammatory cytokines within both the brain and periphery (Fig. 4b, Extended Data Fig 3b). We therefore measured the weight of mice prior to sacrifice, detecting a significant reduction in the weight of monocyte engrafted mice (Fig. 6h). Despite this, monocyte and microglia engrafted mice exhibited equivalent motor function via accelerating Rotarod (Fig. 6i). However, in two other naturalistic tasks: marble burying and nest-building, monocyte engrafted mice exhibited greatly reduced activity (Fig. 6j,k). Low levels of nest building and burying activity have previously been shown to provide a sensitive measure of health and depression-like states and can be induced by systemic LPS treatment. 41,42 Therefore, to determine whether the observed behavioral impairments might be associated with reductions in weight, we performed a correlation analysis. This comparison revealed strong positive correlations between weight and both nest building (r=0.82, *P<0.013) and marble burying (r=0.83, *P<0.012), suggesting that the observed reductions in weight and naturalistic behavior in monocyte engrafted mice provide evidence of sickness behavior. Collectively, our behavioral analyses strongly indicate that widespread engraftment of human monocytes leads to significant neurobehavioral dysfunction. Discussion Microglia arise early in development from HPCs within the yolk sac that give rise to primitive macrophages that migrate into the developing CNS. 11,12 In contrast, blood monocytes and most other peripheral myeloid cells arise later in development and continue to be produced throughout life from definitive bone marrow-derived HSPCs. 11,12 Lineage tracing studies have further confirmed that the CNS receives little-to-no input from definitive HSPC-derived monocytes or macrophages in healthy mice. 10 Nevertheless, recent studies have coupled transplantation of definitive HSPCs, myeloid progenitors, or monocytes with microglia depleting treatments, to replace endogenous microglia with peripherally derived murine myeloid cells. 1-5 These initial studies have shown exciting evidence of reduced pathology and improved outcomes in mouse models of AD, LSD, and FTD. However, these along with other studies have also shown that engrafted MLCs remain transcriptionally and functionally distinct from microglia. 6-9 As human and murine microglia also exhibit many species-dependent differences, 8,13-16 the potential impact of human MLCs on brain function remains unclear. It is therefore critically important to investigate the effects of human MLC engraftment in appropriate in vivo systems before proceeding to clinical translation. The current study together with the accompanying report by Wang et. al., independently investigated the differences between human iPSC-microglia and peripherally derived monocytes following transplantation into the brains of adult microglia-deficient xenotolerant FIRE mice. Four months post-transplantation, despite similar numbers of engrafted cells and widespread brain distribution, we detected many important differences in the transcriptional and functional properties of brain engrafted monocytes versus microglia. Specifically, we find that human monocytes, but not microglia, led to a chronic proinflammatory state characterized by pronounced astrogliosis, demyelination, and neuronal and synaptic loss, culminating in severe behavioral impairments. Collectively, our data demonstrate that ontogeny plays a critical role in the long-term function of myeloid cells within the brain, suggesting that clinical development of approaches that promote the replacement of microglia with definitive myeloid cells instead of primitive-like iPSC-microglia warrants considerable caution. A major strength of our findings is the remarkable similarities between our current results and the accompanying report by Wang and colleagues. Each study was conceived, pursued, and interpreted independently and yet comparisons between the two datasets reveal extremely strong concordance (Fig. 2h, Extended Data Fig. 1b,c). Importantly, the two studies also utilized complimentary approaches to examine the impact of myeloid cell ontogeny on brain health. Whereas Wang et.al., utilized epigenetic approaches to further understand the transcriptional regulation of definitive versus primitive myeloid ells, we utilized proteomic and behavioral approaches to further examine the downstream consequences of these differing ontogenies. Despite these robust and consistent results, it is important to acknowledge some potential limitations of our study. To achieve brain wide engraftment of human monocytes in a model system we utilized immune deficient mice that lack, B, and NK cells. As a result, potential interactions between adaptive immunity and innate myeloid cells within the brain cannot be examined. The future development of more complex humanized models that employ matching definitive and primitive immune cell engraftment could provide a promising approach to address this shortcoming. Another potential caveat to our approach is that we examined the differences between complete brain wide monocyte versus microglia engraftment. Clinical approaches would be unlikely to fully replace endogenous microglia and thus understanding the potential interactions between engrafted monocytes and endogenous microglia will likely be informative. For example, can proinflammatory monocytes alter the activation state of endogenous microglia or vice versa? Future studies that examine co-transplantation of human microglia and monocytes could therefore be highly informative. In conclusion, the current study demonstrates the importance of human myeloid cell ontogeny on neuronal and glial health and behavioral function. Together with the accompanying study by Wang et. al., our findings suggest that CNS wide replacement of microglia with definitive human myeloid cells should be examined with considerable caution and that the use of stem cell derived microglia that better mimic the primitive ontogeny of microglia may provide a safer alternative. METHODS Animals All animal procedures were conducted in accordance with guidelines set forth by the National Institutes of Health and the University of California, Irvine Institutional Animal Care and Use Committee. Original CSF1R ΔFIRE/ΔFIRE mice were generated and previously characterized by Clare Pridans and David Hume. 43 FIRE mice were generated on a B6CBAF1/J background. Heterozygous FIRE mice were then crossed with wild-type C57BL/6J mice for four generations before backcrossing for four generations with xenotransplantation-compatible MITRG mice (Jackson Laboratories # 017711) which harbor deletions in Rag2 and IL2rγ and humanized CSF1, CSF2/Il3 and THPO. The resulting mice were maintained via breeding of heterozygous FIRE females (Fire -/+ , Rag2 -/- , il2rg -/- , hCSF1 +/+ , hCSF2/Il3 +/+ , hTHPO +/+ ) with homozygous (Fire -/- , Rag2 -/- , il2rg -/y , hCSF1 +/+ , hCSF2/Il3 +/+ , hTHPO +/+ ) or heterozygous males (Fire -/+ , Rag2 -/- , il2rg -/y , hCSF1 +/+ , hCSF2/Il3 +/+ , hTHPO +/+ ). Genotyping of FIRE mice was performed using the following primers: mCSF1R_FIRE_F (5’-GCGGTTGTAGGAAACCCTGA-3’), mCSF1R_WT_F (5’-GGTGCCAGCAATGTGTTTCC-3’), and mCSF1R_R: (5’-CACTCCTACCACTGGGCATC-3’). All mice were age- and sex-matched and group-housed, but single-housed post-transplantation, on a 12-h/12-h light/dark cycle with food and water ad libitum . Mice were housed with ambient temperature and humidity. Cages and bedding were changed every 1–2 w. Behavioral analysis utilized hFIRE mice, 19 a highly similar xenotolerant model that lacks murine microglia but does not include humanized CSF2 and THPO. Cell lines Peripheral Blood negatively sorted CD14 + monocytes from four male subjects were obtained from Lonza (#4W-400; see details below). Four iPSC lines from healthy male subjects were generated and obtained from the Alzheimer’s Disease Research Core (ADRC) at University of California Irvine (see details in the table). Informed consent was received under an approved institutional IRB and Non-integrating Sendai virus (Life Technology, Cytotune 2.0) used to perform iPSC reprogramming. iPSCs clones were then confirmed to exhibit normal karyotype by Array Comparative Genomic Hybridization (ACGh by Cell Line Genomics), exhibit pluripotency via trilineage in vitro differentiation (Trilineage Differentiation Kit, Stem Cell Technologies), and sterility via MycoAlert (Lonza). Sample Cell Type Line ID Sex Age (at donation) Race Source 1 CD14+ mono 26198 Male 23 Caucasian Lonza 2 CD14+ mono 23617 Male 45 Caucasian Lonza 3 CD14+ mono 23623 Male 29 Unknown Lonza 4 CD14+ mono 29631 Male 49 Caucasian Lonza 5 iPSC ADRC 75 Male 64 Caucasian UCI ADRC 6 iPSC ADRC 76 Male 83 Caucasian UCI ADRC 7 iPSC ADRC 77 Male 77 Caucasian UCI ADRC 8 iPSC ADRC 40 Male 66 Asian UCI ADRC Microglia differentiation from iPSCs iPSC-derived microglia (iMG) were differentiated using our previously published and highly replicated protocol. 44 To begin HPC differentiation, iPSCs were passaged onto 6-well matrigel-coated plates in TeSR-E8 at a density of 80 colonies of 100 cells each per 35 mm well. On day 0, cells were transferred to Medium A from the STEMdiff Hematopoietic Kit (05310; STEMCELL Technologies). On day 3, cells were exposed to Medium B and remained in Medium B for 7 additional days while small round HPCs began to lift off the colonies. On day 10, non-adherent CD43+ HPCs were collected by carefully removing medium and non-adherent cells with a serological pipette. At this point, HPCs were frozen in BamBanker freezing solution (Wako, NC9582225) for long-term storage. For iMG differentiation, HPCs were cultured on 6-well matrigel-coated plates in complete iMG medium (DMEM/F12 [11039-021; Thermo Fisher Scientific], 2× insulin-transferrin-selenite [41400045; Thermo Fisher Scientific], 2× B27 [17504044; Thermo Fisher Scientific], 0.5× N2 [17502048; Thermo Fisher Scientific], 1× Glutamax [35050-061; Thermo Fisher Scientific], 1× non-essential amino acids [11140-050; Thermo Fisher Scientific], 400 mM monothioglycerol, and 5 mg/mL human insulin [I2643; Sigma-Aldrich] freshly supplemented with 100 ng/mL IL-34 [200-34; Peprotech], 50 ng/mL TGFb1 [100-21; Peprotech], and 25 ng/mL M-CSF [300-25; Peprotech]) for 28 days. Fresh complete iMG media was added every 24 h. iMG were frozen in BamBanker freezing solution (Wako, NC9582225) for long-term storage. Cells used for iMG transplantation were later thawed in complete iMG medium for 18–24 h to recover before being resuspended at 62,500 cells/µl in 1× Dulbecco's Phosphate Buffered Saline (DPBS; low Ca2+, low Mg2+) for transplantation. A total of 500K cells (Monocyte or iMG) were injected per mouse. Adult Intracranial Transplantation All mouse surgeries and use were performed in strict accordance with approved National Institutes of Health (NIH) and institutional guidelines. Direct bilateral intracranial injections were performed as previously described. 15,45 Briefly, adult mice (2 m old) were anesthetized under continuous isoflurane and secured to a stereotaxic frame (Kopf), and local anesthetic (Lidocaine 2%; 17478-711-31, Medline) was applied to the head before exposing the skull. Using a 30-gauge needle affixed to a 10-mL Hamilton syringe, mice received 2 μL of cell suspension in sterile 1× DPBS (14190144; Thermo Fisher Scientific) at 125,000 cells/μL at each injection site. Monocyte and iMG transplantations were conducted bilaterally into the lateral parietal association cortex and dorsal hippocampus at the following coordinates relative to bregma: anteroposterior, 2.06 mm; mediolateral, ± 1.75 mm; dorsoventral, 1.75 mm (hippocampus) and 0.95 mm (cortex). Cells were injected at a rate of 62,500 cells/30 s with 4 min diffusion time in between injections. The needle was cleaned with consecutive washes of DPBS, 70% (vol/vol) ethanol, and DPBS in between hemispheres and animals. Animals were allowed to recover on heating pads before being placed in their home cages and received 2 mg/mL Acetaminophen (0904-7014-16; Major Pharmaceuticals) diluted in water for 5 d. Collection and Isolation of Peripheral Blood Plasma Blood was collected into an EDTA-anticoagulant collection tube and immediately placed on ice. Tubes were then transferred to a pre-chilled centrifuge and spun at 1,600 rcf for 15 min at 4°C. Following centrifugation, plasma supernatant was gently collected and transferred to a 1.5 mL Eppendorf tube on ice. Eppendorf tubes were then transferred to a pre-chilled centrifuge and spun at 1,600 rcf for 15 min at 4°C. Following the second centrifugation, plasma supernatant was collected and aliquoted into 30 μL sub-aliquots and kept at -80°C for long-term storage. Isolation of Soluble Protein Homogenates Animals were perfused (20 mL per min) with 1X DPBS (4°C) for 3 min. Brains were surgically excised from each mouse and cut in half along the mid-sagittal plane. The left hemisphere was placed in 4% (wt/vol) paraformaldehyde (PFA) for 36 h at 4°C for subsequent immunohistochemistry (IHC) and spatial transcriptomics and the right hemisphere fresh-frozen on dry ice and stored at -80°C for biochemical analysis and proteomics. Fresh frozen hemispheres were crushed on liquid nitrogen using mortar and pestle and were split into 2 aliquots. One aliquot was used for Proteomics. The second aliquot was homogenized in solution of T-PER (78510, Thermo Scientific, Waltham, MA) and phosphatase and protease inhibitor mixtures (78426 and 78426, Thermo Scientific, MA). Homogenates were centrifuged at 16,000 rcf for 30 min at 4°C and supernatants were stored at −80°C for analysis. Protein concentration of soluble brain samples was determined using BCA Protein Assay Kit (Pierce) and 2 mg/mL of total protein per sample was run for ELISAs. Quantitative Biochemical Analysis Analysis of soluble brain samples were conducted using commercially available ELISAs following manufacturer guidelines to measure Glial Fibrillary Acidic Protein (ABIN6574131; antibodies-online.com), Synaptophysin (ABIN6959799; antibodies-online.com), Postsynaptic density protein 95 (K15OQND; MSD), human XRCC5/Ku80 (MBS7608104; MyBioSource.com); human complement C4 (ab108824; Abcam); mouse IL-34 (M3400; R&D systems) and V-PLEX Proinflammatory Panel 1 Human Kit (K15049D; MSD). Analysis of peripheral blood plasma was also conducted using the V-PLEX Proinflammatory Panel 1 Human Kit (K15049D; MSD). Protein levels for each sample were then calculated via comparison to each assay-specific standard curve. Immunohistochemistry Animals were perfused with DPBS and isolated brain hemispheres were drop fixed in 4% (wt/vol) PFA for 36 h then cryoprotected in a 30% (wt/vol) sucrose at 4°C. Brains were sectioned coronally into 30-μm-thick slices on a freezing microtome (Leica, SM 2010R) and stored in a solution of 0.05% NaN3 (S2002; Sigma-Aldrich) in 1× PBS (P44017-100TAB; Sigma-Aldrich) as free-floating slices. For staining, tissue was blocked for 1 h in 1× PBS, 0.2% Triton X-100 (9002-93-1; Thermo Fisher Scientific), and 10% donkey serum (NC9624464; Thermo Fisher Scientific). Immediately following blocking, brain sections were placed in primary antibodies diluted in 1× PBS and 1% donkey serum and incubated ON whilst shaking at 4°C. Samples were then incubated in conjugated secondary antibodies for 1 h followed by mounting on microscope slides. Sections were labeled with combinations of goat anti-IBA1 (1:300; ab5076; Abcam), rabbit anti-IBA1 (1:300; 019-19741; Wako), rabbit anti-Ku80 (1:200; ab80592; Abcam), mouse anti-Ku80 (1:250; MAS-12933; Invitrogen), rabbit anti-P2RY12 (1:500; HPA014518; Millipore Sigma), and chicken anti-GFAP (1:2000; ab4674; Abcam) overnight at 4°C followed by 1 h incubation at 23°C whilst shaking. After washing with 1× PBS for 5 m three times, sections were incubated with highly cross-adsorbed AlexaFluor-conjugated secondary antibodies (1:400; ThermoFischer) for 1 h in the dark, then washed three times with 1× PBS before mounting with Fluoromount-G (0100-01; SouthernBiotech). Imaging acquisition and processing Immunofluorescent sections were visualized and captured using an Olympus FX3000 confocal microscope using identical confocal and Z-stack settings for a given quantitative comparison. Half brain stitches to capture half-brains with high resolution were performed using Fluoview FV31S-DT software. For each quantitative analysis, three sections were imaged per region per animal using the “Surfaces” function in IMARIS software (Bitplane). All images were examined blinded using batch processing. Numbers, area, volume, or intensity were calculated for each section, and these values were averaged to provide a single value per animal for subsequent statistical comparisons. For some representative images, brightness and contrast settings were slightly and equally adjusted across groups to reveal fine structures and morphology. Importantly, no changes were made to any images used for quantification. Spatial transcriptomics and analysis Fixed half brains were cryoprotected in 30% sucrose and then sectioned in the coronal plane with a freezing sliding microtome. 10µm thick sections at the level of the dorsal hippocampus (Bregma AP -2.06) were collected and mounted on Superfrost plus glass microscope slides and stored at 4°C overnight to dry. Slide preparation was performed following the 10X Visium CytAssist Spatial protocol (Rev A). Each slide contained two capture areas, and each capture area was designed to hold three adjacent half brain coronal sections, allowing for parallel spatial profiling of multiple experimental conditions within a single assay. Sections were stained with hematoxylin and eosin (H&E) and imaged using brightfield microscopy to capture histological architecture. Spatial transcriptomic libraries were prepared using the 10x Genomics Human Transcriptome Probe Set v2, which uses probe hybridization and ligation-based capture (rather than poly-A enrichment), making it suitable for RNA derived from Paraformaldehyde fixed tissue. Although this probe set is designed to detect human transcripts, expression of genes such as GFAP, AQP4, and SNAP25—typically associated with murine astrocytes and neurons—was also observed. These signals were not computationally removed. Instead, it was assumed that myeloid-associated transcripts were derived from the engrafted human cells, as the host FIRE mouse lacks murine microglia. Library preparation, probe hybridization, and amplification were carried out by the UCI Genomics Core Facility, following the manufacturer’s protocols. Sequencing was performed on an Illumina NovaSeq 6000 platform using paired-end reads (28 bp for read 1 and 91 bp for read 2), achieving a target sequencing depth of 50,000–100,000 reads per spot, consistent with recommended parameters for FFPE tissue. Each Visium spot represents pseudobulk gene expression from approximately 1 to 10 cells, depending on local tissue density and spot coverage. Tissue Demultiplexing and Brain Region Annotation Each capture area included three distinct brain tissue sections, representing either biological replicates or different experimental conditions. To demultiplex the spatial data, manual segmentation was performed in Loupe Browser v8.0 or higher (10x Genomics). Using the aligned H&E-stained image, each tissue section was visually outlined based on its physical boundaries, and only barcodes falling entirely within each outlined area were extracted for downstream analysis. There was no spatial overlap between the three brain tissue sections within each capture area, ensuring that each barcode could be unambiguously assigned to a single tissue section. Spots located outside the H&E-detected tissue area were excluded from the analysis. To annotate anatomical brain structures, each tissue section was manually assigned to one of 13 predefined brain regions, including upper and lower cortical layers, the hippocampus, striatum, thalamus, amygdala, and white matter tracts (Fig 2a). These assignments were guided by visual alignment to reference coronal sections from the Allen Mouse Brain Atlas, using anatomical landmarks such as ventricles, white matter tracts, and cortical boundaries. The resulting barcode-to-region assignments were added to the Seurat object as metadata and were used in region-stratified analyses, including differential gene expression, module score comparisons, and spatial visualization. Data Preprocessing and Quality Control Sequencing reads were processed using Space Ranger v3.1.0 with the appropriate GRCh38 reference genome and FFPE Human Transcriptome Probe Set v2 annotation files (10x Genomics). Raw outputs—including feature-barcode matrices, spatial image data, and tissue position files—were imported into R (v4.4.2) using the Load10X_Spatial() function from Seurat v5.3.0. Given the fragmented nature of RNA in FFPE tissue, a rigorous quality control (QC) pipeline was applied to retain only high-quality barcodes corresponding to biologically informative tissue spots. Global distributions of key QC metrics were visualized using observed quantile plots, allowing empirical determination of filtering thresholds. The following filters were applied: Total UMI count (nCount_Spatial) > 800, Number of detected genes (nFeature_Spatial) > 300, Percentage of mitochondrial transcripts 0.05. The complexity score was calculated as the ratio of unique genes to total UMI per spot: complexity = nFeature_Spatial / nCount_Spatial. This metric captures transcript diversity relative to sequencing depth and helped identify and exclude low-complexity (potentially damaged or ambient RNA–dominated) barcodes. Barcodes outside the tissue area, as defined by H&E-based segmentation, were excluded during demultiplexing. Each spot was annotated with sample ID, experimental condition, slide ID, and capture area as metadata. Slide- and region-specific barcode prefixes were standardized to enable consistent downstream merging and integration. Normalization, Integration, and Dimensionality Reduction Each sample was independently normalized using SCTransform, which stabilizes variance while adjusting for technical confounders such as sequencing depth and mitochondrial gene content. To account for batch effects due to differences in slide, capture area, and processing time, data were integrated using Harmony, which aligns datasets in principal component (PC) space. Slide ID and capture area were specified as covariates for batch correction, enabling alignment of biological signals across technical replicates while preserving spatial and anatomical structure. Dimensionality reduction was performed using Principal Component Analysis (PCA) followed by Uniform Manifold Approximation and Projection (UMAP) for two-dimensional visualization. Graph-based clustering was performed using Seurat’s FindNeighbors() and FindClusters() functions with resolutions ranging from 0.4 to 0.8. Clusters were assessed for biological granularity and stability across resolutions. A resolution of 0.6 was selected for downstream analyses, as it produced a non-redundant but sufficiently detailed set of clusters. This choice was informed by cluster tree analysis and best practices outlined in the Bioconductor OSCA workflow (https://bioconductor.org/books/3.12/OSCA/clustering.html). Spatial feature plots were generated using Seurat functions and maintained precise registration to H&E histological images, preserving anatomical context for downstream spatial analyses. Differential Gene Expression Analysis To quantify transcriptional differences across conditions and regions, we applied two complementary frameworks—spot-level testing and pseudobulk aggregation—to retain spatial resolution while adding replicate-aware statistical control to limit false positives. Spot-level differential expression was performed on SCT-normalized data using Seurat’s FindMarkers() (Wilcoxon rank-sum), including batch covariates in the model. Significance was defined as Benjamini–Hochberg–adjusted P (FDR) 1. This approach is sensitive to spatially localized or cell-type–specific differences but treats spots as independent replicates, which can inflate false discoveries due to within-sample correlation—a recognized limitation in single-cell/spatial analyses. 46,47 Pseudobulk differential expression was also performed to incorporate biological replication and mitigate pseudo-replication, raw counts were aggregated per sample and analyzed with edgeR v3.42.4 by contrasting Monocyte condition to iMG. Libraries were normalized using TMM, and GLMs were fit with condition and batch covariates. Genes were called significant at FDR 1. This replicate-aware framework improves control of false discovery and detects global, consistent shifts across samples. Combining both spot-level and pseudobulk frameworks balances spatial sensitivity (spot-level) with statistical robustness (pseudobulk). We prioritized high-confidence DEGs detected by both methods; spot-only DEGs were interpreted as localized/cell-type–restricted signals representing spatially resolved transcriptional programs, whereas pseudobulk-only DEGs reflected broad but modest sample-level/condition specific shifts. This dual strategy follows best-practice recommendations for spatial/single-cell DE. 46 Module Score Calculation Gene program activity per spatial transcriptomics spot was quantified using Seurat's AddModuleScore() function, applied to SCT-normalized expression values. We curated three transcriptional modules to capture distinct myeloid activation states relevant to the xenotransplant context: 1) Disease-Associated Microglia (DAM), 2) Homeostatic microglia ( P2RY12, P2RY13, CX3CR1, ACY3 , MX1 ) and 3) Monocytes ( CD14, MSR1, LGALS1, GPNMB, ITGAX ). Gene sets were manually curated based on prior literature and pathway relevance. Spatial-seq versus RNA-seq Conmparisons Genes were plotted in a ratio–ratio plot and points were colored according to significance and directionality thresholds. Overall concordance between datasets was quantified using the Pearson correlation coefficient. Venn plots comparing upregulated monocyte genes across the two studies compared genes that reached significance in both studies (Log2FC>1, FDR<0.05) via hypergeometric P. Visualization and Statistical Reporting Intersection (Venn) analysis : Overlap between spot-level and pseudobulk DEGs was summarized with VennDiagram with ggven reporting counts and percentages for each set and the intersection. Genes in the intersection were prioritized as high-confidence significant results for downstream summaries and figures. Volcano plots : Differential expression results were visualized with EnhancedVolcano using the pseudobulk result as background, displaying log₂FC vs. –log₁₀(FDR). The significant genes (intersection of the two methods) were colored coded by group (iMG vs. Monocyte), with selected marker genes labeled using ggrepel to avoid overlap. Violin plots with statistics: Expression of selected genes and module scores was summarized with violin plots (ggplot2), showing median and interquartile ranges. Two-sided Wilcoxon rank-sum tests were performed using ggpubr::stat_compar e_means , with Benjamini–Hochberg correction applied when testing multiple genes/regions. Adjusted P values were displayed directly on the plots. Spatial maps genes and modules: Spatial distributions of single genes and module scores (DAM, IFN, iMG, Monocyte) were rendered with Seurat’s SpatialFeaturePlot and SpatialDimPlot , and, where helpful, customized using SeuratExtend/scCustomize for consistent legends and image ordering across samples. All spatial plots were registered to H&E images. Proteomics Tandem mass tag (TMT)–labeled brain tissue lysates was processed as previously described. 48,49 Samples (100ug) were reduced and alkylated with 5 mM DTT and 10mM IAA in the dark for 30 min. The samples were then digested with 4ug of Lysyl endopeptidase (Wako) overnight. Samples were then diluted with 7X 100mM Tris with 4ug trypsin (ThermoFisher) and digested overnight at room temperature. The peptide solutions were acidified to a final concentration of 1% (vol/vol) formic acid (FA) and 0.1% (vol/vol) trifluoroacetic acid (TFA) then mixed 1:1 with 4% H3PO4 and desalted with a 10 mg Prime MXC plate (Water's). Each MCX column was loaded directly after acidification. The wells were then washed with 500ul of L 100 mM NH4COOH / 2% formic acid and then with 500ul of methanol. Samples were eluted with 500ul of (50:50 ACN:MeOH + 5% NH4OH) and immediately dried by a SpeedVac. For TMT labeling, each brain peptide digest was resuspended in 75 µL of 100 mM triethylammonium bicarbonate (TEAB) buffer, and 5 mg of TMTpro reagents was dissolved in 200 µL ACN. Then 100 µg of peptide samples were aliquoted and resuspended in 100 µL TEAB buffer. After bringing the TMTpro reagents to room temperature and mixing with ACN, 41 µL of the TMT solution was added to each peptide solution and incubated for 1 h. The reaction was stopped with the addition of 8 µL of 5% hydroxylamine. The labeled samples were then combined, concentrated using a SpeedVac, and diluted with 0.1% TFA. After acidification, the peptides were desalted using a C18 Sep‐Pak column, washed, and eluted with 50% ACN. Each sample was assessed for complete labeling on a single-shot LC-MS/MS analysis prior high pH fractionation. High-pH Off-line Fractionation Dried samples were re-suspended in high pH loading buffer (0.07% vol/vol NH4OH, 0.045% vol/vol FA, 2% vol/vol ACN) and loaded onto a Water’s BEH 1.7 um 2.1mm by 150mm. A Thermo Vanquish was used to carry out the fractionation. Solvent A consisted of 0.0175% (vol/vol) NH4OH, 0.01125% (vol/vol) FA, and 2% (vol/vol) ACN; solvent B consisted of 0.0175% (vol/vol) NH4OH, 0.01125% (vol/vol) FA, and 90% (vol/vol) ACN. The sample elution was performed over a 25 min gradient with a flow rate of 0.6 mL/min. A total of 192 individual equal volume fractions were collected across the gradient and subsequently pooled by concatenation into 96 fractions and dried to completeness using a SpeedVac. Liquid Chromatography Each sample was resuspended in 10 ul of loading buffer (0.1% FA) and 2 ul was analyzed by liquid chromatography coupled to tandem mass spectrometry. Peptide eluents were separated on Water’s CSH 150um x 15cm column packed with 1.7um resin. Buffer A was water with 0.1% (vol/vol) formic acid, and buffer B was 80% (vol/vol) acetonitrile in water with 0.1% (vol/vol) formic acid. Elution was performed over a 30 min gradient. The non-linear gradient was from 1% to 99% solvent B. Mass Spectrometry Peptides were monitored by Orbitrap Eclipse spectrometer (ThermoFisher) fitted with a high-field asymmetric waveform ion mobility spectrometry (FAIMS Pro) ion mobility source (ThermoFisher). Two compensation voltages (CV) of -45 and -65 were chosen for FAIMS. Each 1.5 second cycle consisted of one full scan (MS1) was performed with an m/z range of 410-1600 at 60,000 resolution, 100% AGC and 50 ms injection time and higher energy collision-induced dissociation (HCD) TMT scans collected at 30,000 resolutions with Turbo TMT on. The mass range was set to start at 120 m/z, isolation was set to 0.7 D and AGC was set to 250% with a max injection time of 86 ms. Database Search The raw files were searched using FragPipe (version 19.1). The FragPipe pipeline relies on MSFragger (version 3.7) 50,51 for peptide identification and Philosopher (version 4.8.0) 52 for FDR filtering and downstream processing. The chimeric Uniprot protein database used here contains canonical isoforms for both Mouse (Mus musculus; accessed 02/2023) and Human (Homo sapiens, accessed 02/2019). The workflow used in FragPipe followed default TMT-16 plex parameters, used for both TMT-16 and TMT-18 experimental design. Briefly, precursor mass tolerance was -20 to 20 ppm, fragment mass tolerance of 20 ppm, mass calibration and parameter optimization were selected, and isotope error was set to -1/0/1/2/3. Enzyme specificity was set to strict-trypsin and up to two missing trypsin cleavages were allowed. Peptide length was allowed to range from 7 to 50 and peptide mass from either 200 to 5,000 Da. Variable modifications that were allowed in our search included: oxidation on methionine, N-terminal acetylation on protein, and N-terminal acetylation on peptide, with a maximum of 3 variable modifications per peptide. Peptide Spectral Matches were validated using Percolator. 53 The false discovery rate (FDR) threshold was set to 1% and protein and peptide abundances were quantified using Philosopher for downstream analysis. Protein Quantitation and Quality Control The protein abundances are first normalized by scaling total protein signal within each channel for each specific sample to the maximum channel-specific total signal. We then used a tunable median polish approach, TAMPOR, to remove technical batch variance in the proteomic data, as previously described 54 TAMPOR is utilized to remove intra-batch and inter-batch variance while preserving meaningful biological variance in protein abundance values, normalizing to the median of selected intra-batch samples. This approach is robust to outliers and columns with up to 50% values missing. If a protein had more than 50% samples with missing values, it was removed from the matrix. No imputation of missing values was performed for any cohort. For the current data, TAMPOR leverages the median protein abundance from the pooled Global Internal Standard (GIS) TMT channels as the denominators in both factors to normalize sample-specific protein abundances across the 2 batches of samples. Proteomic data processing and analysis Normalized abundances were log₂-transformed prior to modeling. Differential abundance modeling: Differential protein abundance was assessed in R (v4.4.2) using limma (3.62.2) with empirical Bayes moderation. Condition Monocyte versus iMG was modeled as the primary effect; when multiple samples derived from the same donor were present, within-donor correlation was estimated with duplicateCorrelation(). Significance was defined as FDR (Benjamini–Hochberg) 0.5 or > 1 as indicated. Quality control and visualization: Global data structure and sample comparability were evaluated by principal component analysis (PCA) and hierarchical clustering. Heatmaps were generated with pheatmap (1.0.13 using row-wise z-scores of log₂ abundances and annotated by condition. Volcano plots were produced with EnhancedVolcano (1.24.0) to display log fold change versus significance. Violin plots: For each protein, log2- transformed abundances were visualized as violin plots using ggplot2. Violin bodies show kernel density estimates; centered boxplots display the median and interquartile range. Proteins were ordered by the difference in mean expression between conditions (Monocyte − iMG). For each gene, differences between iMG and Monocyte were assessed with a two-sided, unpaired Wilcoxon rank-sum test ( wilcox.test via ggpubr::compare_means ). The resulting p-values are shown in Figure 3c. MoSeq 3D depth videos were recorded in a circular 17” diameter open-field arena (OFA) using a Microsoft Axure Kinect camera. Before recording, the mice were habituated in the examination room for 45 min. During the behavioral recording, the mice were freely moving in the arena for 30 minutes, then put back to its home cape. Between recording sessions, the area was cleaned according to the Moseq protocol. The recorded videos were analyzed using the MoSeq framework, available at http://moseq4all.org. MoSeq is an unsupervised machine learning framework that process and analyze animal spontaneous behavior into sub-second, repeated and modulated segments called “syllables”. The pipeline first processed the video such that the mouse was in the center of the video and the mouse was pointing right. The processed videos then went through dimensionality reduction and modeling to get the frame-by-frame behavioral labels. Syllable usages were aggregated by experimental group in the analysis. Syllable usage differences were identified using Kruskal-Wallis test and a post hoc Dunn’s two-sided test with permutation, using Bejamini-Hochberg FDR (α=0.05). 55 Linear discriminant analysis (LDA) is a method that finds the linear combination of inputs to maximize group separation. To visualize the behavioral difference between the xenografted mice, LDA took syllable usages and experimental group-day pair as input to learn the low dimensional embedding. The position heatmap was computed by aggregating the mouse centroid positions and the heatmap represents the percentages of time. Distance traveled was computed by summing the between frame Euclidean distance between the mouse centroids in two consecutive frames within a given duration. Marble Burying Standard polycarbonate mouse cages were fitted with filter-top covers and filled with 5mm of fresh, unscented bedding material. 15 glass marbles (3 rows, equally spaced, with 5 marbles per row) were placed at the surface of the bedding. Mice in their home cages were then transferred to the behavioral room to acclimate. After 45 min, each mouse was transferred to the behavioral cage. After 30 min, mice were transferred back to their individual home cages, and the number of buried marbles (>50% below the surface of the bedding) were scored by three blinded observers as previously described 56 whose scores were averaged per mouse. Nesting Mice were transferred to a standard polycarbonate mouse cage with filter-top cover containing a single nestlet. After 24 hours, nest building was scored on a scale of 1-5 as previously described 57 by three blinded observers whose scores were averaged per mouse per day. Mice were then transferred to a new cage with a new nestlet and similarly scored a second time after 24 hours. Mice were then transferred to a new cage for the third time with a new nestlet and similarly scored after 24 hours. Final scores were calculated by averaging nesting score across three days per mouse. Statistic Calculations GraphPad Prism 9 was used to perform statistical tests and generate P values. We used standard designation of P values throughout the Figures (ns = not significant or p ≥0.05; * p <0.05; ** p <0.01; *** p <0.001; ****p<0.0001). Values are represented as average mean ± SEM. Details of number of replicates and the specific statistical test used are provided in the individual figure legends. Data represented for all antibodies calculated from an average of 2-3 matched coronal sections per animal. For quantification of NeuN density a Grubbs outlier test identified one statistical outlier which was removed. Because group distributions of NeuN counts deviated from normality, we utilized a non-parametric Kruskal–Wallis test followed by Dunn’s multiple comparisons. Declarations Data Availability The spatial RNA-seq datasets are available through the GEO Super Series accession number (will be added prior to publication). Proteomic data will be made available via Further information and requests for resources and reagents should be directed to and will be fulfilled by Mathew Blurton-Jones ( [email protected] ). Acknowledgements We thank Drs. David Hume and Clare Pridans for generously providing the original immune intact FIRE-knockout mouse model. iPSC lines were generated by the UCI-ADRC iPS cell core funded by NIH P30AG066519. Figures 1a, 1b, 3a, and 6a were created with Biorender (https://www.biorender.com). Author contributions: Conceptualization, H.D., JP.C., M.B.-J. Methodology, H.D., JP.C., G.E.-S., S.L., J.H., A.L.C., J.K.C., D.D., J.N., C.T., S.K.S., F.W., A.S., K.M., M.M., T.E.L., A.K.M., L.L., A.L. Quantitative analysis, JP.C., H.D., A.L. Spatial sequencing preparation and analysis, G.E.-S, J.K.C., J.H., A.L. Proteomic data collection and analysis, D.D., F.W, A.S., N.T.S., J.H., A.L. Writing, JP.C., H.D. and M.B.-J. Writing revisions, JP.C., H.D. M.B.-J., A.L., J.H., S.L., S.R.D., N.T.S. Supervision, M.B.-J; N.T.S; S.R.D. Funding acquisition, M.B.-J, JP.C., H.D., J.H., A.L.C., N.T.S., S.R.D., E.H. Funding This work was supported in part by the Larry L. Hillblom Foundation #2025-A-175-FEL, NIH T32 AG000096, and NIH T32 AG073088 (to JP.C.); NIH T32 AG000096 (to J.H.); Larry L. Hillblom Foundation #2024-A-023-FEL and BrightFocus Postdoctoral Fellowship Program in Alzheimer's #A2025007F (to G.E.-S); CIRM EDUC4-12822 (to A.L.C.); NIH R01 AG061895 (to H.D.); NIH P30AG066519 (to E.H.); NIH R01AG073625 (to S.R.D); NIH U01AG061357 (N.T.S.); and NIH RF1NS139975, NIH R01AG082255, NIH U19AG069701, NIH P30AG066519, Alzheimer’s Association ADSF-25-1459986-C, and the Cure Alzheimer’s Fund Neuroimmune Consortium (to M.B.-J.). Generation of the ADRC iPSC lines were performed by the UCI-ADRC iPSC core funded by NIH P30AG16573. Competing Interests JP.C, H.D., J.H. and M.B.-J. are co-inventors on patent applications filed by the University of California Regents related to genetic modification of cells to confer resistance to CSF1R antagonists (WO2022/212897), transplantation of stem cell-derived microglia to treat leukodystrophies (WO2024/015933), and pathology-responsive recombinant cells (WO2023/212663). M.B.-J. is a co-founder and consultant for Savanna Biotherapeutics. No other disclosures were reported. References Yoo, Y., Neumayer, G., Shibuya, Y., Mader, M. M. & Wernig, M. A cell therapy approach to restore microglial Trem2 function in a mouse model of Alzheimer's disease. Cell Stem Cell 30 , 1043-1053.e1046 (2023). https://doi.org/10.1016/j.stem.2023.07.006 P, C. et al. CNS-wide repopulation by hematopoietic-derived microglia-like cells corrects progranulin deficiency in mice - PubMed. Nature communications 15 (07/05/2024). https://doi.org/10.1038/s41467-024-49908-4 Shibuya, Y. et al. Treatment of a genetic brain disease by CNS-wide microglia replacement. Sci Transl Med 14 , eabl9945 (2022). https://doi.org/10.1126/scitranslmed.abl9945 Frosch, M. et al. Microglia–neuron crosstalk through Hex–GM2–MGL2 maintains brain homeostasis. Nature 2025 (2025-08-06). https://doi.org/10.1038/s41586-025-09477-y Mader, M. M.-D. et al. Therapeutic genetic restoration through allogeneic brain microglia replacement. Nature 2025 (2025-08-06). https://doi.org/10.1038/s41586-025-09461-6 Cronk, J. C. et al. Peripherally derived macrophages can engraft the brain independent of irradiation and maintain an identity distinct from microglia. J Exp Med 215 , 1627-1647 (2018). https://doi.org/10.1084/jem.20180247 Shemer, A. et al. Engrafted parenchymal brain macrophages differ from microglia in transcriptome, chromatin landscape and response to challenge. Nat Commun 9 , 5206 (2018). https://doi.org/10.1038/s41467-018-07548-5 J, B. et al. Monocytes can efficiently replace all brain macrophages and fetal liver monocytes can generate bona fide SALL1+ microglia - PubMed. Immunity 58 (05/13/2025). https://doi.org/10.1016/j.immuni.2025.04.006 FC, B. et al. A Combination of Ontogeny and CNS Environment Establishes Microglial Identity - PubMed. Neuron 98 (06/27/2018). https://doi.org/10.1016/j.neuron.2018.05.014 A, S. et al. Dual ontogeny of disease-associated microglia and disease inflammatory macrophages in aging and neurodegeneration - PubMed. Immunity 55 (08/09/2022). https://doi.org/10.1016/j.immuni.2022.07.004 B, A., JL, B., C, K., KM, M. & FM, R. Infiltrating monocytes trigger EAE progression, but do not contribute to the resident microglia pool - PubMed. Nature neuroscience 14 (07/31/2011). https://doi.org/10.1038/nn.2887 Mildner, A. et al. Microglia in the adult brain arise from Ly-6ChiCCR2+ monocytes only under defined host conditions. Nature Neuroscience 2007 10:12 10 (2007-11-18). https://doi.org/10.1038/nn2015 D, G. et al. An environment-dependent transcriptional network specifies human microglia identity - PubMed. Science (New York, N.Y.) 356 (06/23/2017). https://doi.org/10.1126/science.aal3222 TF, G. et al. Transcriptomic analysis of purified human cortical microglia reveals age-associated changes - PubMed. Nature neuroscience 20 (2017 Aug). https://doi.org/10.1038/nn.4597 Hasselmann, J. et al. Development of a Chimeric Model to Study and Manipulate Human Microglia In Vivo. Neuron 103 , 1016-1033.e1010 (2019). https://doi.org/10.1016/j.neuron.2019.07.002 Lund, H. et al. Competitive repopulation of an empty microglial niche yields functionally distinct subsets of microglia-like cells. Nat Commun 9 , 4845 (2018). https://doi.org/10.1038/s41467-018-07295-7 Abud, E. M. et al. iPSC-Derived Human Microglia-like Cells to Study Neurological Diseases. Neuron 94 , 278-293.e279 (2017). https://doi.org/10.1016/j.neuron.2017.03.042 CZ, H. et al. Human microglia maturation is underpinned by specific gene regulatory networks - PubMed. Immunity 56 (09/12/2023). https://doi.org/10.1016/j.immuni.2023.07.016 Chadarevian, J. P. et al. Therapeutic potential of human microglia transplantation in a chimeric model of CSF1R-related leukoencephalopathy. Neuron (2024). https://doi.org/10.1016/j.neuron.2024.05.023 Keren-Shaul, H. et al. A Unique Microglia Type Associated with Restricting Development of Alzheimer's Disease. Cell 169 , 1276-1290.e1217 (2017). https://doi.org/10.1016/j.cell.2017.05.018 McQuade, A. et al. Gene expression and functional deficits underlie TREM2-knockout microglia responses in human models of Alzheimer's disease. Nat Commun 11 , 5370 (2020). https://doi.org/10.1038/s41467-020-19227-5 R, M. et al. Xenografted human microglia display diverse transcriptomic states in response to Alzheimer's disease-related amyloid-β pathology - PubMed. Nature neuroscience 27 (2024 May). https://doi.org/10.1038/s41593-024-01600-y Chadarevian, J. et al. Harnessing human iPSC-microglia for CNS-wide delivery of disease-modifying proteins - PubMed. Cell stem cell (2025). https://doi.org/10.1016/j.stem.2025.03.009 D, K., T, U. & T, T. Functions and development of red pulp macrophages - PubMed. Microbiology and immunology 59 (2015 Feb). https://doi.org/10.1111/1348-0421.12228 L, D.-O. et al. FEAST: A flow cytometry-based toolkit for interrogating microglial engulfment of synaptic and myelin proteins - PubMed. Nature communications 14 (09/27/2023). https://doi.org/10.1038/s41467-023-41448-7 JM, S. et al. Lipofuscin-like autofluorescence within microglia and its impact on studying microglial engulfment - PubMed. Nature communications 14 (11/03/2023). https://doi.org/10.1038/s41467-023-42809-y U, E. & AV, M. Defining microglial-synapse interactions - PubMed. Science (New York, N.Y.) 381 (09/15/2023). https://doi.org/10.1126/science.adh7906 Claes, C. et al. Plaque-associated human microglia accumulate lipid droplets in a chimeric model of Alzheimer's disease. Mol Neurodegener 16 , 50 (2021). https://doi.org/10.1186/s13024-021-00473-0 MS, H. et al. APOE4/4 is linked to damaging lipid droplets in Alzheimer's disease microglia - PubMed. Nature 628 (2024 Apr). https://doi.org/10.1038/s41586-024-07185-7 A, L. et al. Amelioration of Tau and ApoE4-linked glial lipid accumulation and neurodegeneration with an LXR agonist - PubMed. Neuron 112 (02/07/2024). https://doi.org/10.1016/j.neuron.2023.10.023 MM, C. et al. Steatosis drives monocyte-derived macrophage accumulation in human metabolic dysfunction-associated fatty liver disease - PubMed. JHEP reports : innovation in hepatology 5 (08/11/2023). https://doi.org/10.1016/j.jhepr.2023.100877 NR, W. & KJ, M. Macrophage-derived foam cells in atherosclerosis: lessons from murine models and implications for therapy - PubMed. Current drug targets 8 (2007 Dec). https://doi.org/10.2174/138945007783220597 M, v. E. & JMFG, A. The Unique Phenotype of Lipid-Laden Macrophages - PubMed. International journal of molecular sciences 22 (04/14/2021). https://doi.org/10.3390/ijms22084039 Wade, C., Runeckles, K., Chataway, J., Houlden, H. & Lynch, D. S. CSF1R-Related Disorder: Prevalence of CSF1R Variants and Their Clinical Significance in the UK Population. Neurology: Genetics 10 (2024 Jul 16). https://doi.org/10.1212/NXG.0000000000200179 Dulski, J. et al. Global Presence and Penetrance of CSF1R-Related Disorder. Neurology: Genetics 10 (2024 Sep 13). https://doi.org/10.1212/NXG.0000000000200187 Kempthorne, L. et al. Loss of homeostatic microglial phenotype in CSF1R-related Leukoencephalopathy. Acta Neuropathol Commun 8 , 72 (2020). https://doi.org/10.1186/s40478-020-00947-0 Dulski, J. et al. Hematopoietic Stem Cell Transplantation in. Pharmaceutics 14 (2022). https://doi.org/10.3390/pharmaceutics14122778 Du, S. et al. Mutations in the human CSF1R gene impact microglia’s maintenance of brain white matter integrity. Nature Immunology 2025 26:7 26 (2025-06-26). https://doi.org/10.1038/s41590-025-02195-7 AB, W. et al. Mapping Sub-Second Structure in Mouse Behavior - PubMed. Neuron 88 (12/16/2015). https://doi.org/10.1016/j.neuron.2015.11.031 D, K., E, M. & N, I. Monitoring Pathogen-Induced Sickness in Mice and Rats - PubMed. Current protocols in mouse biology 7 (06/19/2017). https://doi.org/10.1002/cpmo.27 Tagawa, N., Tsuneoka, Y. & Funato, H. Nest-building behavior in laboratory mice: Multifunctional roles and neural mechanisms. Neuroscience & Biobehavioral Reviews 178 (2025/11/01). https://doi.org/10.1016/j.neubiorev.2025.106377 Jiang, X. et al. Proanthocyanidin prevents lipopolysaccharide-induced depressive-like behavior in mice via neuroinflammatory pathway. Brain Research Bulletin 135 (2017/10/01). https://doi.org/10.1016/j.brainresbull.2017.09.010 Rojo, R. et al. Deletion of a Csf1r enhancer selectively impacts CSF1R expression and development of tissue macrophage populations. Nat Commun 10 , 3215 (2019). https://doi.org/10.1038/s41467-019-11053-8 McQuade, A. et al. Development and validation of a simplified method to generate human microglia from pluripotent stem cells. Mol Neurodegener 13 , 67 (2018). https://doi.org/10.1186/s13024-018-0297-x Chadarevian, J. P., Davtyan, H., Lombroso, S. I., Bennett, F. C. & Blurton-Jones, M. CRISPR generation of CSF1R-G795A human microglia for robust microglia replacement in a chimeric mouse model. STAR Protoc 4 , 102490 (2023). https://doi.org/10.1016/j.xpro.2023.102490 Squair, J. W. et al. Confronting false discoveries in single-cell differential expression. Nature Communications 2021 12:1 12 (2021-09-28). https://doi.org/10.1038/s41467-021-25960-2 Zimmerman, K. D. et al. A practical solution to pseudoreplication bias in single-cell studies. Nature Communications 2021 12:1 12 (2021-02-02). https://doi.org/10.1038/s41467-021-21038-1 Seifar, F. et al. Large-scale deep proteomic analysis in Alzheimer's disease brain regions across race and ethnicity. Alzheimers Dement 20 , 8878-8897 (2024). https://doi.org/10.1002/alz.14360 Higginbotham, L. et al. Unbiased classification of the elderly human brain proteome resolves distinct clinical and pathophysiological subtypes of cognitive impairment. Neurobiol Dis 186 , 106286 (2023). https://doi.org/10.1016/j.nbd.2023.106286 AT, K., FV, L., DM, A., D, M. & AI, N. MSFragger: ultrafast and comprehensive peptide identification in mass spectrometry-based proteomics - PubMed. Nature methods 14 (2017 May). https://doi.org/10.1038/nmeth.4256 F, Y. et al. Identification of modified peptides using localization-aware open search - PubMed. Nature communications 11 (08/13/2020). https://doi.org/10.1038/s41467-020-17921-y da Veiga Leprevost, F. et al. Philosopher: a versatile toolkit for shotgun proteomics data analysis. Nature Methods 2020 17:9 17 (2020-07-15). https://doi.org/10.1038/s41592-020-0912-y Käll, L. et al. Semi-supervised learning for peptide identification from shotgun proteomics datasets. Nature Methods 2007 4:11 4 (2007-10-21). https://doi.org/10.1038/nmeth1113 EB, D., NT, S. & ECB, J. Batch Correction and Harmonization of -Omics Datasets with a Tunable Median Polish of Ratio - PubMed. Frontiers in systems biology 3 (2023). https://doi.org/10.3389/fsysb.2023.1092341 AB, W. et al. Revealing the structure of pharmacobehavioral space through motion sequencing - PubMed. Nature neuroscience 23 (2020 Nov). https://doi.org/10.1038/s41593-020-00706-3 Witkin, J. M. Animal Models of Obsessive‐Compulsive Disorder. Current Protocols in Neuroscience 45 (2008/10/01). https://doi.org/10.1002/0471142301.ns0930s45 Deacon, R. M. & Deacon, R. M. Assessing nest building in mice. Nature Protocols 2006 1:3 1 (2006-08-24). https://doi.org/10.1038/nprot.2006.170 Additional Declarations Yes there is potential Competing Interest. JP.C, H.D., J.H. and M.B.-J. are co-inventors on patent applications filed by the University of California Regents related to genetic modification of cells to confer resistance to CSF1R antagonists (WO2022/212897), transplantation of stem cell-derived microglia to treat leukodystrophies (WO2024/015933), and pathology-responsive recombinant cells (WO2023/212663). M.B.-J. is a co-founder and consultant for Savanna Biotherapeutics. No other disclosures were reported. Supplementary Files DavtyanTable1DEGs.xlsx Table 1 DavtyanTable2DEPs.xlsx Table 2 DavtyanTable3KEGGenrichment.xlsx Table 3 ExtendedDataFigures.docx Cite Share Download PDF Status: Under Review 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-7762816","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Biological Sciences - Article","associatedPublications":[],"authors":[{"id":523834201,"identity":"671f06a7-be35-423d-8418-b91ce23e0095","order_by":0,"name":"Mathew Blurton-Jones","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAsElEQVRIiWNgGAWjYDACCeYGMM0PxAcYGJiJ0cII1JLAwCDZQLIWgwNgLhFadGc3tklX/rDJM76R/PAAQ4V1YgMhLWZ3DrZJnklIKza7kQa06Ew6EVpuJLZJNiQcTtx2I4fhAGPbYRK0bJ4B0vKPFC0bJEBaGojRcudgs2VDWlrijDPPDA4kHEs3JqzldvPBmw02Non97cmPP3yosZYlqAUVJJCmfBSMglEwCkYBLgAALfxFwWnzfJsAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-7770-7157","institution":"University of California, Irvine","correspondingAuthor":true,"prefix":"","firstName":"Mathew","middleName":"","lastName":"Blurton-Jones","suffix":""},{"id":523834202,"identity":"fc8b167f-7832-47da-8e7b-b5ac87d1af95","order_by":1,"name":"Hayk Davtyan","email":"","orcid":"https://orcid.org/0000-0001-8554-6456","institution":"University of California Irvine","correspondingAuthor":false,"prefix":"","firstName":"Hayk","middleName":"","lastName":"Davtyan","suffix":""},{"id":523834203,"identity":"119e9e7e-834c-4ab4-a220-9863f47f846a","order_by":2,"name":"Jean Paul Chadarevian","email":"","orcid":"https://orcid.org/0000-0002-2478-6592","institution":"University of California Irvine","correspondingAuthor":false,"prefix":"","firstName":"Jean","middleName":"Paul","lastName":"Chadarevian","suffix":""},{"id":523834204,"identity":"2c0381dd-a393-4c16-bef3-b8444b8e1afc","order_by":3,"name":"Anita Lakatos","email":"","orcid":"","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Anita","middleName":"","lastName":"Lakatos","suffix":""},{"id":523834205,"identity":"c99eae30-d823-408c-b9f4-50ffdddeb986","order_by":4,"name":"Sherry Lin","email":"","orcid":"https://orcid.org/0000-0002-1275-8391","institution":"Harvard Medical School","correspondingAuthor":false,"prefix":"","firstName":"Sherry","middleName":"","lastName":"Lin","suffix":""},{"id":523834206,"identity":"f0a64c77-b137-4690-8ac4-96dd496bc11d","order_by":5,"name":"Jonathan Hasselmann","email":"","orcid":"https://orcid.org/0000-0003-2630-9850","institution":"University of California Irvine","correspondingAuthor":false,"prefix":"","firstName":"Jonathan","middleName":"","lastName":"Hasselmann","suffix":""},{"id":523834207,"identity":"ff690341-ebe3-4ba6-9b43-7ef5727cc97f","order_by":6,"name":"Jasmine Nguyen","email":"","orcid":"","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Jasmine","middleName":"","lastName":"Nguyen","suffix":""},{"id":523834208,"identity":"20ecd27a-ca77-4aa4-926b-b58e2bc84276","order_by":7,"name":"Duc Duong","email":"","orcid":"","institution":"Emory University","correspondingAuthor":false,"prefix":"","firstName":"Duc","middleName":"","lastName":"Duong","suffix":""},{"id":523834209,"identity":"0cbeea9b-8a15-455a-a68f-ac78ffe1e9d0","order_by":8,"name":"Joia Capocchi","email":"","orcid":"","institution":"University of California Irvine","correspondingAuthor":false,"prefix":"","firstName":"Joia","middleName":"","lastName":"Capocchi","suffix":""},{"id":523834210,"identity":"6b6b96f4-e433-40c8-8da2-d721c5367b42","order_by":9,"name":"Alina Lahian","email":"","orcid":"","institution":"University of California Irvine","correspondingAuthor":false,"prefix":"","firstName":"Alina","middleName":"","lastName":"Lahian","suffix":""},{"id":523834211,"identity":"3b6318d4-976b-4ce4-bd68-a693ca084397","order_by":10,"name":"Kimiya Mansour","email":"","orcid":"","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Kimiya","middleName":"","lastName":"Mansour","suffix":""},{"id":523834212,"identity":"89100394-b832-4474-af18-ad5699c2e404","order_by":11,"name":"Ghazaleh Eskandari-Sedighi","email":"","orcid":"https://orcid.org/0000-0003-0447-7722","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Ghazaleh","middleName":"","lastName":"Eskandari-Sedighi","suffix":""},{"id":523834213,"identity":"8a747d21-4daa-4d22-9d6f-1c07de66b00e","order_by":12,"name":"Christina Tu","email":"","orcid":"","institution":"University of California Irvine","correspondingAuthor":false,"prefix":"","firstName":"Christina","middleName":"","lastName":"Tu","suffix":""},{"id":523834214,"identity":"e1c964b1-4687-4054-8192-c678dd47d1b1","order_by":13,"name":"Sepideh Kiani Shabestari","email":"","orcid":"","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Sepideh","middleName":"Kiani","lastName":"Shabestari","suffix":""},{"id":523834215,"identity":"b53d184b-86a0-4d40-b160-d3743c71c7b2","order_by":14,"name":"Fang Wu","email":"","orcid":"","institution":"Emory University","correspondingAuthor":false,"prefix":"","firstName":"Fang","middleName":"","lastName":"Wu","suffix":""},{"id":523834216,"identity":"14fd83af-25fe-4165-9bbb-09602bf0a52d","order_by":15,"name":"Anantharaman Shantaraman","email":"","orcid":"","institution":"Emory University","correspondingAuthor":false,"prefix":"","firstName":"Anantharaman","middleName":"","lastName":"Shantaraman","suffix":""},{"id":523834217,"identity":"03e2fe89-1098-4222-827d-f0c7f88916eb","order_by":16,"name":"Michael Mgerian","email":"","orcid":"","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"","lastName":"Mgerian","suffix":""},{"id":523834218,"identity":"fe9c51bf-7ebf-4c02-932f-b5134f7300cb","order_by":17,"name":"Tau Lim","email":"","orcid":"","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Tau","middleName":"","lastName":"Lim","suffix":""},{"id":523834219,"identity":"437f7c5d-5e82-4c9a-bbaa-49e7c2ea0b00","order_by":18,"name":"Alan Mai","email":"","orcid":"","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Alan","middleName":"","lastName":"Mai","suffix":""},{"id":523834220,"identity":"7f6c403f-13a6-467e-acf3-3b69470cb81e","order_by":19,"name":"Lauren Le","email":"","orcid":"https://orcid.org/0009-0006-8527-4554","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Lauren","middleName":"","lastName":"Le","suffix":""},{"id":523834221,"identity":"2cd67bab-b86b-4302-a85f-834df6cd3b7c","order_by":20,"name":"Ani Agababian","email":"","orcid":"","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Ani","middleName":"","lastName":"Agababian","suffix":""},{"id":523834222,"identity":"1e34229b-eabb-46fa-98e3-aee0356f71fb","order_by":21,"name":"Elizabeth Head","email":"","orcid":"","institution":"University of California","correspondingAuthor":false,"prefix":"","firstName":"Elizabeth","middleName":"","lastName":"Head","suffix":""},{"id":523834223,"identity":"9b7287fa-52cc-40bc-a8b2-b1349cc88b0a","order_by":22,"name":"Sandeep Robert Datta","email":"","orcid":"https://orcid.org/0000-0002-8068-3862","institution":"Harvard Medical School","correspondingAuthor":false,"prefix":"","firstName":"Sandeep","middleName":"Robert","lastName":"Datta","suffix":""},{"id":523834224,"identity":"5078baf5-2d52-4d52-acb4-b2473aceadc7","order_by":23,"name":"Nicholas Seyfried","email":"","orcid":"https://orcid.org/0000-0002-4507-624X","institution":"Emory University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Nicholas","middleName":"","lastName":"Seyfried","suffix":""}],"badges":[],"createdAt":"2025-10-02 00:40:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7762816/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7762816/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93204933,"identity":"5a4e548b-53ee-4dbb-83ef-a846853dd2f5","added_by":"auto","created_at":"2025-10-10 07:49:52","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1444503,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHuman iPSC-microglia and blood-derived monocytes fully engraft within the brains of adult hFIRE mice but adopt differing morphologies. (a) \u003c/strong\u003eMicroglia arise from primitive hematopoietic stem cells within the yolk sac, whereas monocytes arise later in development from bone marrow derived definitive hematopoietic stem and progenitor cells. Lineage tracing studies have shown that very few monocytes infiltrate the healthy brain, but increased numbers migrate into the brain in response to injury, irradiation-induced loss of microglia, or neurodegeneration. \u003cstrong\u003e(b) \u003c/strong\u003eSchematic depicting transplantation paradigm of 2-month-old xenotolerant FIRE mice with iPSC-microglia (Mg) or human CD14+ monocytes (Mo) derived from 4 male subjects then sacrificed at 5.5-months of age. \u003cstrong\u003e(c-d) \u003c/strong\u003eBrain wide human (Ku80, human-specific nuclei, white) xenografted iMG (xMG) and monocytes (IBA1, green), scale bar: 500µm.\u003cstrong\u003e (e-f) \u003c/strong\u003eQuantification of xenografted microglia and monocytes in half brain coronal sections of xenotolerant hFIRE mice; \u003cem\u003ep\u003c/em\u003e values from unpaired, two-sided t test. Half brain (Bregma AP -1.94): t(16)= 0.6068; ns=0.5525. Forebrain (Bregma AP +0.14): t(16)= 0.01849; ns=0.9855.\u003cstrong\u003e (g-h) \u003c/strong\u003eRepresentative confocal microscopy combined with IMARIS-based branching analysis of xenografted microglia and monocytes, scale bar: 20µm.\u003cstrong\u003e (i-j) \u003c/strong\u003eQuantification of xenografted microglia and monocytes branch complexity within the cortex and hippocamps; \u003cem\u003ep\u003c/em\u003e values from unpaired, two-sided t test (Cortex: t(16)= 8.087; ****\u003cem\u003eP\u003c/em\u003e\u0026lt;0.0001) (Hippocampus: t(16)=3.446; **\u003cem\u003eP\u003c/em\u003e=0.0033). Data represented as average mean value ± SEM (Mg, n = 7; Mo, n = 11). ns, not significant; ∗∗p \u0026lt; 0.01, and ∗∗∗∗p \u0026lt; 0.0001.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-7762816/v1/4cca5c697f43e8e8a50f8266.png"},{"id":93204934,"identity":"0ad9697f-f45e-4df5-adfe-69f3194a6047","added_by":"auto","created_at":"2025-10-10 07:49:52","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1722262,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSpatial sequencing of monocyte and microglia engrafted mice reveal numerous differences in myeloid cell activation state.\u003c/strong\u003e \u003cstrong\u003e(a)\u003c/strong\u003e Representative graph-based cluster identification from spot level (17,799 spots) computed from 10 principal components using Louvain clustering algorithm and color-coded brain segmentation. \u003cstrong\u003e(b) \u003c/strong\u003eUniform manifold approximation and projection (UMAP) plot of analyzed spots from the brains of FIRE mice xenografted with human microglia or monocytes. \u003cstrong\u003e(c) \u003c/strong\u003e1768 Differentially Expressed Genes (DEGs) are identified from the overlap between Spot-based and pseudobulk analysis. \u003cstrong\u003e(d) \u003c/strong\u003eVolcano plot comparing overlapping DEGs from xenografted brain sections, reveals numerous significant alterations between microglia and monocyte engrafted brains (Log\u003csub\u003e2\u003c/sub\u003e(FC) ≥ 1; FDR ≤ 0.05). \u003cstrong\u003e(e) \u003c/strong\u003e\u003cem\u003eP2Ry12\u003c/em\u003e, \u003cem\u003eAPOE\u003c/em\u003e, and \u003cem\u003eCD74\u003c/em\u003e mRNA expression (normalized and scaled expression) between engrafted human microglia and monocytes. \u003cstrong\u003e(f)\u003c/strong\u003e Expression level of \u003cem\u003eACY3, CSF1R, CX3CR1, P2RY12, CD14, APOE, BIN1, LIPA, GPNMB, CD9, CD74, CTSD, LGALS3, MS4A7, ITGAX, MAFB, IRF7, STAT1, MX1, and CIITA\u003c/em\u003e in brain-wide xenografted human microglia and monocytes. \u003cstrong\u003e(g) \u003c/strong\u003eModule scores of the disease-associated microglia (DAM) signature (444 upregulated genes) across segmented brain regions. \u003cem\u003eP \u003c/em\u003evalues in f and g were calculated by two-sided Wilcoxon test and adjusted using the Benjamini–Hochberg method. \u003cstrong\u003e(h)\u003c/strong\u003e Ratio–Ratio plot comparing the Log2Fc differences between microglia and monocytes from spatial sequencing data in comparison to scSeq data from the accompanying study by Wang et. al. Concordance between datasets was quantified via Pearson correlation coefficient (r = 0.51, p \u0026lt; 2 x 10\u003csup\u003e-16\u003c/sup\u003e).\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-7762816/v1/a8e0f1efe3597229ce235eb5.png"},{"id":93204939,"identity":"d4ef6679-622f-4d94-9c0b-eb64ba25e84b","added_by":"auto","created_at":"2025-10-10 07:49:52","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1820502,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProteomic analysis reveals increased levels of many proinflammatory proteins in monocyte engrafted brains.\u003c/strong\u003e \u003cstrong\u003e(a)\u003c/strong\u003eSchematic depicting frozen hemibrain tissue and protein processing for TMT Mass Spec Analysis (n=8 mice per group, engrafted with n=4 Monocyte or Microglia lines). \u003cstrong\u003e(b) \u003c/strong\u003eHeatmap showing log2-transformed intensities of the top 70 differentially expressed proteins (DEPs) between microglia and monocytes engrafted brains (n=8 mice/group). \u003cstrong\u003e(c) \u003c/strong\u003eExpression levels of P2RY12, LGALS3, CD14, GPNMB, CX3CR1, CD74, ITGAX, OAS2, ACY3, CTSD, MX1, CD9, APOE, IFIT3, BIN1, and CSF1R protein. \u003cstrong\u003e(d)\u003c/strong\u003e \u0026nbsp;Principal component analysis reveals microglia or monocyte identity as the primary source of variation between xenografted brains. \u003cstrong\u003e(e) \u003c/strong\u003eRepresentative immunostaining for LGALS3 positive (blue) microglia or monocytes (IBA1, green) in the cortex of xenografted mice; scale bar: 40µm. \u003cstrong\u003e(f-g) \u003c/strong\u003eQuantification of LGALS3 within the cortex and hippocampus, \u003cem\u003ep\u003c/em\u003e values from unpaired, two-sided t test (Cortex t(16)= 7.568; ****\u003cem\u003eP\u003c/em\u003e\u0026lt;0.0001) (Hippocampus: t(16)=4.571; ***\u003cem\u003eP\u003c/em\u003e=0.0003). \u003cstrong\u003e(h-k) \u003c/strong\u003eRepresentative immunostaining for human nuclei (Ku80, white), P2RY12 (red), GPNMB (purple), HLA-DR (orange), and CD9 (pastel red) positive microglia or monocytes (IBA1, green) in the cortex of xenografted mice; scale bar: 40µm. Quantification of P2RY12 \u003cstrong\u003e(I,m)\u003c/strong\u003e, GPNMB \u003cstrong\u003e(n,o)\u003c/strong\u003e, HLA-DR \u003cstrong\u003e(p,q)\u003c/strong\u003e, and CD9 \u003cstrong\u003e(r,s)\u003c/strong\u003e immunoreactivity in the cortex and hippocampus of xenografted mice, \u003cem\u003ep\u003c/em\u003e values from unpaired, two-sided t test (P2RY12 Cortex: t(16)= 20.83; ****\u003cem\u003eP\u0026lt;\u003c/em\u003e0.0001) (P2RY12 Hippocampus t(16)= 9.605; ****\u003cem\u003eP\u003c/em\u003e\u0026lt;0.0001) (GPNMB Cortex: t(16)= 2.904; *\u003cem\u003eP\u003c/em\u003e=0.0104) (GPNMB Hippocampus t(16)= 2.424; *\u003cem\u003eP\u003c/em\u003e=0.0276) (HLA-DR Cortex: t(16)= 7.592; ****\u003cem\u003eP\u003c/em\u003e\u0026lt;0.0001) (HLA-DR Hippocampus t(16)= 6.299; ****\u003cem\u003eP\u0026lt;\u003c/em\u003e0.0001) (CD9 Cortex: t(16)= 5.448 ****\u003cem\u003eP\u003c/em\u003e\u0026lt;0.0001) (CD9 Hippocampus t(16)= 4.079; ***\u003cem\u003eP\u003c/em\u003e=0.009). Data represented as average mean value ± SEM. Immunohistological analysis examined n=7 microglia and n=11 monocyte engrafted mice. ns, not significant; ∗p \u0026lt; 0.05, ∗∗p \u0026lt; 0.01, ∗∗∗p \u0026lt; 0.001, and ∗∗∗∗p \u0026lt; 0.0001.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-7762816/v1/8629087f7637b861613581e4.png"},{"id":93205989,"identity":"094dcca7-6b9e-483a-aac7-07acaeece0dd","added_by":"auto","created_at":"2025-10-10 07:57:52","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2091192,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProteomic analysis reveals consistent reductions in neuronal, synaptic, and oligodendroglial proteins in monocyte-engrafted mice. (a)\u003c/strong\u003e Heatmap showing log2-transformed intensities in a subset of 74 non-myeloid DEPs between microglia and monocyte engrafted brains (n=8/group). \u003cstrong\u003e(b)\u003c/strong\u003e ELISA quantification of Synaptophysin (SYP), Post Synaptic Density-95 (PSD-95), Myelin Basic Protein (MBP), Glial Fibrillary Acidic Protein (GFAP), Complement component 4 (C4), Tumor Necrosis Factor-Alpha (TNF-α), Interleukin-1 beta (IL-1β), Interleukin-2 (IL-2), Interleukin-8 (IL-8), Interleukin-12 (IL-12), Interleukin-34 (IL-34), and Interferon gamma (IFN-γ) in soluble half-brain lysates; \u003cem\u003ep \u003c/em\u003evalues from one-way ANOVA (SYP: F(2,28)= 24.65; ****\u003cem\u003ep\u003c/em\u003e\u0026lt;0.0001) (PSD-95: F(2,28)= 3.762; *\u003cem\u003ep\u003c/em\u003e=0.0357) (MBP: F(2,28)= 27.82; ****\u003cem\u003ep\u003c/em\u003e\u0026lt;0.0001) (GFAP: F(2,28)= 18.76; ****\u003cem\u003ep\u003c/em\u003e\u0026lt;0.0001) (C4a: F(2,28)= 26.63; ****\u003cem\u003ep\u003c/em\u003e\u0026lt;0.0001) (TNF-α: F(2,28)= 7.276; **\u003cem\u003ep\u003c/em\u003e=0.0029) (IL-1β: F(2,28)= 77.13; ****\u003cem\u003ep\u003c/em\u003e\u0026lt;0.0001) (IL-2: F(2,28)= 5.828; **\u003cem\u003ep\u003c/em\u003e=0.0077) (IL-8: F(2,28)= 9.766; ***\u003cem\u003ep\u003c/em\u003e=0.0006) (IL-12: F(2,28)= 22.34; ****\u003cem\u003ep\u003c/em\u003e\u0026lt;0.0001) (IL-34: F(2,28)= 111.7; ****\u003cem\u003ep\u003c/em\u003e\u0026lt;0.0001) (IFN-γ: F(2,28)= 7.455; **\u003cem\u003ep\u003c/em\u003e=0.0025) with Tukey multiple comparisons tests. Data represented as average mean value ± SEM (WT, n = 15; Mg, n=8; Mo, n = 8). \u003cstrong\u003e(c)\u003c/strong\u003e Matrix of Pearson correlation (r) coefficients of quantified SYP, PSD-95, MBP, GFAP, C4, TNF-α, IL-1β, IL-2, IL-8, IL-12, IL-34, and IFN-γ. \u003cstrong\u003e(d-e)\u003c/strong\u003e Simple linear regression between levels of SYP and C4 (R\u003csup\u003e2\u003c/sup\u003e\u0026nbsp;= 0.719;\u0026nbsp;****\u003cem\u003eP\u003c/em\u003e\u0026lt;0.0001), SYP and TNF-α (R\u003csup\u003e2\u003c/sup\u003e\u0026nbsp;= 0.577;\u0026nbsp;***\u003cem\u003eP\u003c/em\u003e=0.0006), MBP and C4a (R\u003csup\u003e2\u003c/sup\u003e\u0026nbsp;= 0.558;\u0026nbsp;***\u003cem\u003eP\u003c/em\u003e=0.0009), and MBP and TNF-α (R\u003csup\u003e2\u003c/sup\u003e\u0026nbsp;= 0.4348;\u0026nbsp;**\u003cem\u003eP\u003c/em\u003e=0.0055) in xenografted mice\u003cstrong\u003e. (f)\u003c/strong\u003e Representative confocal imaging of neuronal nuclei (NeuN, green) and xenografted microglia or monocytes (IBA1, green) in the ventral dentate gyrus, scale bar: 40µm. \u003cstrong\u003e(g)\u003c/strong\u003e Quantification of NeuN integrated density within the upper blade of the dentate gyrus of control (WT) or Monocyte (Mo) and Microglia (Mg) engrafted mice. As group distributions of NeuN deviated from normality, we utilized a non-parametric Kruskal–Wallis test (P=0.042) followed by Dunn’s multiple comparisons test, monocyte versus WT mice (Dunn’s: P=0.049), microglia versus WT mice (Dunn’s: P\u0026gt;0.999). \u003cstrong\u003e(h)\u003c/strong\u003e Representative confocal imaging of a neuronal nuclei (NeuN, green) surrounded by phagocytic (CD68, yellow) monocytes (IBA1, green) within the dentate gyrus, scale bar: 10µm. Data represented as average mean value ± SEM (WT, n = 8; Mg, n=6-7; Mo, n = 11). ∗p \u0026lt; 0.05, ∗∗p \u0026lt; 0.01, ∗∗∗p \u0026lt; 0.0001, and ∗∗∗∗p \u0026lt; 0.0001. Comparisons not shown are not significant.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-7762816/v1/22910366758b7bf98f834cf4.png"},{"id":93204935,"identity":"162822e2-9566-411a-8bb8-19fdf123919e","added_by":"auto","created_at":"2025-10-10 07:49:52","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1167339,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe transcriptional signature of brain engrafted monocytes can be detected in human neurodegenerative disease.\u003c/strong\u003e \u003cstrong\u003e(a)\u003c/strong\u003e UMAP plot of homeostatic microglia (blue), proliferating microglia (yellow), ‘ALSP microglia’ (red), border associated macrophages (BAMs, green), and monocytes (purple) from control subjects (N = 10, n = 3,655 nuclei) and \u003cstrong\u003e(d)\u003c/strong\u003epatients with ALSP (N = 5, n=2,650 nuclei) reanalyzed from Du et. al., (Nature Immunology, 2025). The signature of xenografted microglia is enriched within the homeostatic microglia clusters of both control \u003cstrong\u003e(b)\u003c/strong\u003e and to a lesser extent ALSP \u003cstrong\u003e(e)\u003c/strong\u003e subjects. The xenografted monocyte signature is enriched within the monocyte and BAM populations of control subjects \u003cstrong\u003e(c)\u003c/strong\u003e but greatly increased in the ‘ALSP-microglia’ cluster of ALSP patients \u003cstrong\u003e(f)\u003c/strong\u003e. A module score calculated from Du et. al.’s ‘ALSP microglia’ signature is greatly enriched in monocyte engrafted mice \u003cstrong\u003e(g)\u003c/strong\u003e. Correlation analysis of log fold differences between ALSP and control microglia DEGs from Du et. al., in comparison to DEGs from spatial seq analysis of engrafted mice. Module scores generated from Du and colleagues’ ALSP-microglia or homeostatic-microglia signature mapped onto monocyte engrafted \u003cstrong\u003e(i)\u003c/strong\u003e or microglia engrafted \u003cstrong\u003e(j)\u003c/strong\u003e brain sections.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-7762816/v1/3cce0523ddb944b2b63ae515.png"},{"id":93204941,"identity":"ffdaaf6c-0b30-4199-9611-aaeba99331b0","added_by":"auto","created_at":"2025-10-10 07:49:52","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1049294,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBrain wide engraftment of human monocytes promotes significant impairments in behavioral function. (a) \u003c/strong\u003eRepresentative schematic for behavioral testing of xenografted hFIRE mice from 5 to 5.5-months of age. \u003cstrong\u003e(b) \u003c/strong\u003ePosition, Speed, and Moseq syllable usage summary data for microglia (Mg) and monocyte (Mo) xenografted mice over 5 days (n=4 mice/group/day). \u003cstrong\u003e(c) \u003c/strong\u003eMoseq syllable usage of microglia (blue) or monocyte (orange) engrafted mice, stars represent differently used syllables between microglia mice and monocyte engrafted mice (see methods) \u003cstrong\u003e(d) \u003c/strong\u003eLinear Discriminant Analysis (LDA) of Mg or Mo xenografted hFIRE mice over 5 days of recording \u003cstrong\u003e(e)\u003c/strong\u003e Representative summary spatial maps of human microglia or monocyte xenografted mice averaged over five days of recording (top row, larger images) or across individual days (bottom row, small images). \u003cstrong\u003e(f) \u003c/strong\u003eAverage total distance traveled per 5 min intervals of xenografted mice on day 1, Mann-Whitney-Wilcoxon test, *P\u0026lt;0.05. \u003cstrong\u003e(g) \u003c/strong\u003eTotal distance traveled in open field arena over 30 min of recording, Mann-Whitney-Wilcoxon test *P\u0026lt;0.05 \u003cstrong\u003e(h)\u003c/strong\u003e Quantification of average weight of xenografted hFIRE mice at 5.5-months of age, p values from unpaired, two-sided t test (t(6)= 3.080; *P=0.0217). \u003cstrong\u003e(i) \u003c/strong\u003eQuantification of latency to fall off the accelerating rotarod over 5 min of testing, p values from unpaired, two-sided t test (t(6)= 0.3242; ns=0.7568). \u003cstrong\u003e(j) \u003c/strong\u003eAverage number of marbles buried over 30 min of testing, p values from unpaired, two-sided t test (t(6)= 4.387; **P=0.0046). \u003cstrong\u003e(k) \u003c/strong\u003eQuantification of average nesting score over three days of recording, p values from unpaired, two-sided t test (t(6)= 9.971; ****P\u0026lt;0.0001). \u003cstrong\u003e(l) \u003c/strong\u003eCorrelation matrix of behavioral testing of xenografted hFIRE mice. Data represented as average mean value ± SEM (Mg, n = 4; Mo, n = 4). ∗p \u0026lt; 0.05, ∗∗p \u0026lt; 0.01, and ∗∗∗∗p \u0026lt; 0.0001. Comparisons not shown are not significant.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-7762816/v1/fe37b290542a1db6c2fa6d40.png"},{"id":101398236,"identity":"efd2475e-b529-4c0a-9565-17ce249165ea","added_by":"auto","created_at":"2026-01-29 09:40:27","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":10530940,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7762816/v1/1e515b38-2e83-4451-b92a-7b323553fc58.pdf"},{"id":93204940,"identity":"fab4c157-80be-4ebb-8482-dce31dce5bd0","added_by":"auto","created_at":"2025-10-10 07:49:52","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2442041,"visible":true,"origin":"","legend":"\u003cp\u003eTable 1\u003c/p\u003e","description":"","filename":"DavtyanTable1DEGs.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7762816/v1/820b52aef9e234b386c553c7.xlsx"},{"id":93204937,"identity":"b4129bb5-6140-4238-9472-29b48271b427","added_by":"auto","created_at":"2025-10-10 07:49:52","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":115094,"visible":true,"origin":"","legend":"\u003cp\u003eTable 2\u003c/p\u003e","description":"","filename":"DavtyanTable2DEPs.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7762816/v1/7690844feb5ca1b8ebd59272.xlsx"},{"id":93204932,"identity":"0318099c-1b9b-412d-817c-b997e81daccb","added_by":"auto","created_at":"2025-10-10 07:49:51","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":43452,"visible":true,"origin":"","legend":"\u003cp\u003eTable 3\u003c/p\u003e","description":"","filename":"DavtyanTable3KEGGenrichment.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7762816/v1/cc0531187fe6e71ab9f9ac32.xlsx"},{"id":93205990,"identity":"c23907b5-a518-4aff-9994-19a00ca04415","added_by":"auto","created_at":"2025-10-10 07:57:52","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":2018857,"visible":true,"origin":"","legend":"","description":"","filename":"ExtendedDataFigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-7762816/v1/a2d9f41323218515a6329be5.docx"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nJP.C, H.D., J.H. and M.B.-J. are co-inventors on patent applications filed by the University of California Regents related to genetic modification of cells to confer resistance to CSF1R antagonists (WO2022/212897), transplantation of stem cell-derived microglia to treat leukodystrophies (WO2024/015933), and pathology-responsive recombinant cells (WO2023/212663). M.B.-J. is a co-founder and consultant for Savanna Biotherapeutics. No other disclosures were reported.","formattedTitle":"Brain wide engraftment of human monocytes induces neurodegeneration and cognitive dysfunction","fulltext":[{"header":"Main","content":"\u003cp\u003eRecent studies have examined the therapeutic potential of bone marrow transplantation in mouse models of Alzheimer\u0026rsquo;s disease, frontotemporal dementia, and lysosomal storage diseases.\u003csup\u003e1-5\u003c/sup\u003e By combining myeloablative preconditioning with microglial-depleting CSF1R antagonists and hematopoietic stem and progenitor cell (HSPC) transplantation, these studies have shown that brain wide replacement of microglia with murine bone marrow derived \u0026lsquo;microglia-like cells (MLCs) leads to promising reductions in disease-associated pathologies. However, prior studies have also clearly demonstrated that despite long-term brain engraftment, peripherally derived murine MLCs remain transcriptionally, morphologically, and functionally distinct from microglia.\u003csup\u003e6-9\u003c/sup\u003e Given these results along with the well-established differences in microglial versus peripheral myeloid cell ontogeny,\u003csup\u003e10-12\u003c/sup\u003e it is important to understand the long-term consequences of replacing microglia with MLCs. As human and mouse microglia also exhibit significant differences, especially in regard to neurodegeneration-associated genes,\u003csup\u003e8,13-16\u003c/sup\u003e it is equally important to investigate the effects of brain wide human MLC engraftment in appropriate model systems before this approach is translated into clinical practice.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHere we utilized xenotolerant FIRE mice, a mouse model that enables complete humanization of the central nervous system (CNS) microglial compartment, to examine the long-term consequences of replacing endogenous microglia with definitive HSPC-derived human monocytes or induced pluripotent stem cell (iPSC)-derived microglia (iMG) that are generated via a yolk sac-like primitive hematopoietic progenitor cell (HPC) intermediate (Fig. 1a).\u003csup\u003e17\u003c/sup\u003e Fourteen weeks later, brains were examined via transcriptomic, proteomic, biochemical, and histological approaches to determine the impact of myeloid cell ontogeny on brain health and function.\u003c/p\u003e\n\u003cp\u003eTogether with the accompanying study by Wang et. al., we demonstrate robust and reproducible ontogeny-dependent differences in the activation state of definitive HSPC-derived monocytes versus primitive HPC derived microglia. Importantly, these ontogeny-dependent differences mimic those observed in a human neurodegenerative disease and are further accompanied by impairments in neuronal and glial homeostasis, culminating in pronounced cognitive dysfunction. Collectively, our findings suggest that transplantation of iPSC-microglia that closely mimic the primitive developmental ontogeny of human microglia could offer a safer alternative to definitive HSPC transplantation for the treatment of neurological diseases.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHuman monocyte engraftment induces a proinflammatory state\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe previously developed xenotolerant FIRE mice that enable long-term brain wide engraftment of human microglia derived from pluripotent stem cells or surgical resection and demonstrated that engrafted iPSC-derived microglia are highly similar to brain derived human microglia.\u003csup\u003e18,19\u003c/sup\u003e At 2 months of age, male mice were transplanted with human blood monocytes or iMG derived from multiple male control subjects. Fourteen weeks later, brains were examined using a combination of transcriptomic, proteomic, biochemical, and histological approaches (Fig. 1b). Both microglia and monocytes migrated well beyond the hippocampal injection site, leading to similar levels of brain wide engraftment (Fig. 1c-f). Consistent with our prior studies,\u003csup\u003e15,19\u003c/sup\u003e engrafted human iMG adopted a highly ramified cell morphology consistent with the adoption of a homeostatic state (Fig. 1g). While monocytes also exhibited morphological changes with engraftment, branching complexity remained significantly lower than that of microglia and was frequently accompanied by a more elongated morphology (Fig. 1h-j). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo examine the impact of the brain environment on microglial and monocyte gene expression we utilized Visium spatial sequencing. Clustering across all six samples (n=3/group) revealed 13 regionally distinct signatures that corresponded to well established anatomical regions including upper and lower cortical layers, hippocampal subfields, the amygdala, hypothalamus, and white matter tracts (Fig. 2a,b). To maximize rigor, we focused our analysis on the intersection of differentially expressed genes (DEGs) detected using both a spot-based and pseudobulk analyses of the dataset (Fig. 2c). The use of xenotolerant FIRE mice that lack microglia coupled with a human targeted Visium probe set provided additional confidence that the great majority of DEGs detected in this experiment were derived from engrafted human myeloid cells. Consistent with this expectation, pseudobulk comparison between microglia and monocyte-engrafted tissue slices detected 1768 DEGs (LogFc\u0026gt;1, FDR\u0026lt;0.05), many of which represented well-established myeloid genes (Fig. 2d, Supplementary Table\u0026thinsp;1). While microglial engrafted tissue exhibited high levels of canonical homeostatic microglial genes (\u003cem\u003eP2RY12, SALL1, CX3CR1, VSIR\u003c/em\u003e), monocyte-engrafted mice instead exhibited greatly increased expression of genes associated with proinflammatory activation of macrophages such as G\u003cem\u003ePNMB, CD9, APOE, LGALS3, STAT1, CD163,\u0026nbsp;\u003c/em\u003eand \u003cem\u003eCTSD\u003c/em\u003e (Fig. 2d). Spatial maps of \u003cem\u003eP2RY12, APOE,\u0026nbsp;\u003c/em\u003eand\u003cem\u003e\u0026nbsp;CD74\u003c/em\u003e demonstrate the broadly increased expression of \u003cem\u003eP2RY12\u003c/em\u003e in microglial samples in contrast to the greatly increased expression of \u003cem\u003eAPOE\u003c/em\u003e and \u003cem\u003eCD74\u003c/em\u003e in monocyte engrafted brains (Fig. 2e). Spatial mapping of representative genes across all six samples further demonstrated the consistent enrichment of homeostatic microglial genes (\u003cem\u003eP2RY12, SALL1\u003c/em\u003e) versus monocyte/macrophage associated genes (\u003cem\u003eCD14, APOE, CD74, MX1\u003c/em\u003e) across all replicates (Extended Data Fig. 1a). Violin plots of representative DEGs further demonstrate the enrichment of homeostatic genes in microglia and pro-inflammatory genes in monocytes (Fig. 2f, Supplementary Table\u0026thinsp;1). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBrain engrafted monocytes exhibit a DAM-like signature\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe were surprised to find that many of the genes enriched in monocyte engrafted mice (Fig. 2f) corresponded closely with the signature of \u0026lsquo;disease associated microglia (DAM)\u0026rsquo;, a transcriptional state that has consistently been observed in models of Alzheimer\u0026rsquo;s Disease (AD).\u003csup\u003e20\u003c/sup\u003e In previous studies, we and others have similarly reported the induction of a DAM-like signature in human xMG adjacent to beta-amyloid plaques.\u003csup\u003e15,21,22\u003c/sup\u003e In contrast to monocyte engrafted hFIRE mice, xMG engrafted mice in the current study , express low levels of DAM associated genes, and instead exhibit typical homeostatic microglial gene expression (Fig. 2d-f) and morphology (Fig. 1g,i,j). To further understand the potential similarity between the DAM signature and our spatial sequencing data, we used a previously published single cell sequencing (scSeq) dataset from xenografted AD mice\u003csup\u003e15,23\u003c/sup\u003e to generate a DAM module score. Mapping of this module score across each of the 13 brain regions identified in our spatial analysis revealed a significant enrichment of the DAM signature across all regions of the monocyte engrafted brains without evidence of plaque or tangle pathologies in comparison to very low module scores within microglia-engrafted brains (Fig. 2g).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo determine whether our spatial sequencing analysis was consistent with the independently generated scSeq data presented in the accompanying study by Wang and colleagues, we compared the overlap in gene expression between both studies. This analysis revealed a highly significant overlap (P =1.98 x 10\u003csup\u003e-304\u003c/sup\u003e, Extended Data Fig. 1b) with over 40% of upregulated monocyte genes detected by Wang exhibiting a similar upregulation in monocytes within our spatial sequencing dataset. A Ratio-Ratio plot comparing Log2Fc expression differences between monocytes and microglia detected in both studies (Fig. 2h) likewise revealed a strong positive correlation (Pearson r=0.51, P \u0026lt; 2.2 x 10\u003csup\u003e-16\u003c/sup\u003e). To provide a third comparison between these two datasets we generated microglial and monocyte module scores from the Wang et. al. dataset and mapped those scores onto the 13 brain regions examined in our spatial sequencing data. This comparison again revealed a strong overlap between the two datasets across all 13 brain regions (Extended Data Fig. 1c). In summary, the spatial sequencing analysis performed in the current study and the scSeq analysis performed by Wang and colleagues reveal remarkably consistent and reproducible results, especially given the technical differences between the two studies which included differing sequencing approaches, donor cells and sources, mouse colonies, and ages of transplant.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eProteomic analysis further distinguishes inflammatory monocytes from homeostatic microglia\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn addition to our spatial sequencing analysis, rapidly frozen hemibrains from 8 mice per group were homogenized and examined by Tandem Mass Tag mass spectrometry (TMT-MS) (Fig. 3a). Our proteomic analysis revealed 1,406 differentially expressed proteins (DEPs; Adjusted P\u0026lt;0.05) between microglia and monocyte engrafted brains (Supplementary Table\u0026thinsp;2). Of those DEPs, 730 proteins were upregulated in monocyte engrafted brains whereas 676 downregulated DEPs were detected in monocyte engrafted mice (Extended Data Fig. 2a). However, when the top 70 DEPs were compared by heatmap, the great majority of DEGs were increased in monocyte brains, with only 4 out of 70 proteins exhibiting reduced expression in monocyte engrafted brains (Fig. 3b). Consistent with our spatial seq analysis, homoeostatic microglial proteins P2RY12, ACY3, and CRYBB1 exhibited increased expression in microglial engrafted brains, whereas monocyte engrafted brains exhibited greatly increased levels of proinflammatory myeloid and DAM-associated proteins including GPNMB, LGALS3 (Galectin 3), CD9, CD163, FTL (ferritin light chain), LIPA, ITGAX (CD11c) as well as Human leukocyte antigen (HLA) associated proteins involved in both MHC-I (HLA-A, HLA-B, B2M) and MHCII (HLA-DRA, HLA-DRB1, HLA-DPB1, CD74) antigen presentation (Fig. 3b). Violin plots of representative DEPs further demonstrate the consistent enrichment of myeloid activation-associated proteins in monocyte engrafted mice versus increased homeostatic proteins in microglia engrafted brains (Fig. 3c). Principle component analysis of our proteomics data demonstrated the consistency of these findings; revealing a clear separation between the 8 monocyte engrafted samples and 8 microglia engrafted samples along principal component 1 (PC1) representing 44.8% of the variance within the dataset (Fig. 3d).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNext, to further confirm that the expression of myeloid-associated DEPs in half brain homogenates was derived from engrafted human cells, we performed immunohistochemistry (IHC) for five representative DEPs along with IBA-1 to label all myeloid cells and the human-specific nuclear marker KU80 (Fig. 3e-s). IMARIS-based quantitative analysis revealed significantly increased expression of LGALS3 (Galectin-3) within engrafted monocytes in comparison to microglia within both the cortex and hippocampus (Fig. 3e-g). In contrast, IHC analysis of the homeostatic marker P2RY12 demonstrated far greater expression within microglia (Fig. 3h,I,m). Examination of three additional DAM-associated proteins (GPNMB, HLA-DRB1, CD9) also revealed greatly increased expression within human monocytes (Fig. 3 i-k,n-s).\u003c/p\u003e\n\u003cp\u003eTo better define the functional pathways impacted by brain wide monocyte engraftment we performed Gene Ontology (GO) analysis of our proteomic dataset (Supplementary Table\u0026thinsp;3). Representative upregulated GO pathways included terms such as \u0026lsquo;Phagosome, antigen processing presentation, Lipid and atherosclerosis, efferocytosis, TNF signaling, JAK-STAT signaling, Ferroptosis, and cholesterol metabolism\u0026rsquo; (Extended Data Fig. 3b). In contrast, GO pathways that were downregulated in monocyte engrafted mice highlighted terms such as Synaptic vesicle cycle, Cholinergic synapse, GABAergic synapse, Glutamatergic synapse, and Dopaminergic synapse (Extended Data Fig. 3c). Additional downregulated terms included Long-term potentiation (LTP), long-term depression (LTD), neurotrophin signaling, axon guidance, and longevity regulating pathways. As many of these downregulated pathways involve neuronal and synaptic proteins, we next sought to examine the potential impact of monocyte engraftment on non-myeloid cell types.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMonocyte engraftment promotes the loss of neuronal, synaptic, and oligodendroglial proteins\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWhile myeloid associated proteins were highly enriched in the top 70 upregulated DEPs (Figure 3b), our proteomic analysis further revealed a large number of non-myeloid DEPs that were significantly decreased in monocyte engrafted brains (Fig. 4a, Supplementary Table\u0026thinsp;2). Consistent with our analysis of down-regulated GO terms (Extended Data Fig. 2c), many of these non-myeloid DEPs represented canonical neuronal and synaptic proteins such as Tau (MAPT), \u0026beta;3-tubulin (TUBB3), Synaptophysin (SYP), Synapsin-1 (SYN1), Synaptic Vesicle Glycoprotein 2A (SV2A), and PSD-95 (DLG4). Likewise, we observed significant reductions in several canonical oligodendrocyte and oligodendrocyte precursor cell (OPC) proteins including Myelin Basic Protein (MBP), Myelin Oligodendrocyte Glycoprotein (MOG), Myelin-Associated Glycoprotein (MAG), 2\u0026apos;,3\u0026apos;-cyclic nucleotide 3\u0026apos;-phosphodiesterase (CNP), and proteolipid protein 1 (PLP-1) (Fig. 4a). To further confirm these key changes, we next measured levels of murine SYP, PSD-95, and MBP by ELISA. In each case we again observed a significant reduction in these proteins within monocyte engrafted brains (Fig. 4b), whereas microglia-engrafted mice showed no differences in comparison to \u0026lsquo;wild type\u0026rsquo; hCSF1 immune-deficient controls. \u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMonocytes increase astrogliosis and proinflammatory cytokines\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWhile the majority of non-myeloid proteins exhibited decreased expression in monocyte-engrafted brains, several DEPs that are typically enriched in astrocytes or endothelial cells exhibited elevated expression (Fig. 4a, Supplementary Table\u0026thinsp;2). For example, Glial fibrillary acid protein (GFAP), a commonly utilized marker of reactive astrogliosis, along with the astrocytic proteins ALDH1L1, S100\u0026beta;, and aquaporin-4 (AQP4), exhibited elevated expression in monocyte brains (Fig. 4a). ELISA analysis of GFAP and Complement 4 (C4), both of which have frequently been associated with astrogliosis further corroborated our proteomic findings (Fig. 4b). Next, to determine whether the observed increases in myeloid cell activation were further accompanied by changes in cytokine production, we measured ten cytokines via multiplex ELISA. This analysis revealed significantly increased levels of six proinflammatory cytokines within monocyte engrafted brains including TNF\u0026alpha;, Il1\u0026beta;, IL-2, IL-8, IL-12, and INF\u0026gamma;\u0026nbsp;(Fig. 4b), whereas the four remaining analytes (IL-4, IL-6, IL-10, IL-13) showed no significant differences (Extended Data Fig. 3a). Interestingly, IL-34, which along with CSF1 signals via CSF1R to promote microglial differentiation and survival, was likewise elevated within monocyte engrafted brains. Importantly, none of these 10 cytokines were elevated within human microglial engrafted brains in comparison to wildtype hCSF1 immune-deficient controls. To further determine whether there might be a relationship between the observed reductions in synaptic and oligodendroglial proteins and increased proinflammatory cytokines we next generated a correlation table to compare each of the altered brain analytes. This comparison revealed highly significant correlations between many of these analytes including strong inverse associations (blue boxes) between synaptophysin and several proinflammatory cytokines (Fig. 4c, blue boxes). Individual regression analysis of synaptophysin versus C4 and synaptophysin versus TNF\u0026alpha;\u0026nbsp;further revealed highly significant correlations of R\u003csup\u003e2\u003c/sup\u003e=0.719 and R\u003csup\u003e2\u003c/sup\u003e=0.577 respectively (Fig 4d, P\u0026le;0.0006). These same two proinflammatory analytes also exhibited significant inverse relationships with MBP levels (Fig. 4e, P\u0026le;0.0055). \u0026nbsp;Lastly, to determine whether the observed elevations in proinflammatory cytokines can also be detected in the periphery, we performed the same multiplex ELISA on serum samples collected from each mouse at the time of sacrifice. This analysis revealed significant increases in all ten analytes suggesting that monocyte engraftment within the brain also leads to a heightened proinflammatory environment within the periphery (Extended Data Fig. 3b).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHuman monocytes promote hippocampal neurodegeneration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGiven that our proteomic analysis had detected significant decreases in neuronal proteins including Tau and \u0026beta;3-tubulin (Fig. 4a) and our spatial sequencing had revealed highly elevated levels of proinflammatory genes within the hippocampus (Fig. 2e, Extended Data Fig. 1), we performed IHC for the neuronal nuclear marker NeuN and examined the hippocampus. \u0026nbsp;We were surprised to observe a dramatic loss of NeuN+ neurons within the upper blade of the dentate gyrus (Fig. 4f) in 5 of the 11 monocyte-engrafted brains examined. Because group distributions of NeuN deviated from normality, we utilized a non-parametric Kruskal\u0026ndash;Wallis test followed by Dunn\u0026rsquo;s multiple comparisons, revealing a significant reduction in NeuN density in monocyte engrafted mice versus untransplanted \u0026lsquo;wildtype\u0026rsquo; hCSF1 controls (Fig. 4g, Kruskal\u0026ndash;Wallis: P=0.042, \u0026nbsp;Dunn\u0026rsquo;s: P=0.049), but no significant differences between microglia engrafted mice and wildtype controls (Dunn\u0026rsquo;s: P\u0026gt;0.999). Interestingly, co-labeling with IBA-1 revealed densely packed ameboid-shaped monocytes in the same region of the upper blade where NeuN labeling had been lost in monocyte brains. We therefore wondered whether monocytes might be promoting the death and/or clearance of degenerating neurons within this region.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMonocytes, macrophages, and microglia can rapidly phagocytose dying cells. In fact, an important physiological function of monocyte-derived macrophages within the spleen is to phagocytose and degrade aging and dysfunctional red blood cells.\u003csup\u003e24\u003c/sup\u003e However, phagocytosis proceeds rapidly, and antigenicity of internalized proteins is promptly lost by lysosomal degradation, making it challenging to reliably catch snapshots of cellular phagocytosis in the adult brain \u003cem\u003ein vivo\u003c/em\u003e. Three recent reports have further demonstrated the difficulty of reliably quantifying microglial phagocytosis of cells, synapses, and myelin.\u003csup\u003e25-27\u003c/sup\u003e Nevertheless, we observed many examples of NeuN+ neurons within the hippocampus that were surrounded by ameboid-shaped monocytes that expressed high levels of the phagolysosome marker CD68 (Fig. 4h), suggesting that monocytes rapidly clear or perhaps even promote the loss of neurons within the hippocampus. Examples of monocytes with internalized synaptophysin (Extended Data Fig. 4a,b) or myelin basic protein (Extended Data Fig. 4c,d) were also readily observed within the hippocampus and corpus callosum respectively.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhile peptides can be rapidly cleared by the lysosome, a growing number of studies have demonstrated the accumulation and persistence of lipid droplets within microglia during neurodegenerative disease.\u003csup\u003e28-30\u003c/sup\u003e Monocyte-derived macrophages within the periphery are likewise well recognized to accumulate lipid droplets during diseases such as arteriosclerosis, Non-Alcoholic Steatohepatitis, and lysosomal storage diseases.\u003csup\u003e31-33\u003c/sup\u003e We therefore wondered whether brain engrafted monocytes might similarly exhibit lipid droplet accumulation. While IHC analysis of engrafted microglia revealed little to no accumulation of PLIN-2+ lipid droplets, engrafted monocytes exhibited a dramatic increase in lipid droplet density within the cortex and hippocampus (Extended Data Fig. 4e-h). Taken together, our findings suggest that peripherally derived human monocytes adopt a chronically pro-inflammatory signature that promotes the phagocytosis of neurons, synapses, and oligodendrocytes and the accumulation of lipid droplets. Importantly, the accompanying study by Wang and colleagues provides additional independent evidence that human myeloid cell ontogeny strongly influences neuronal and glial health and lipid droplet accumulation. \u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe transcriptional signature of brain engrafted monocytes occurs in human neurodegenerative disease\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCSF1R-related disorder (CSF1R-RD) is a rare neurodegenerative disease associated with mutations in CSF1R, a receptor that is critical for microglia differentiation, survival, and function.\u003csup\u003e34,35\u003c/sup\u003e Also referred to as adult-onset leukoencephalopathy with axonal spheroids and pigmented glia (ALSP), patients affected by this disease exhibit decreased numbers of homeostatic microglia leading to demyelination, astrogliosis, and axonal spheroids which is accompanied by progressive cognitive and/or motor dysfunction.\u003csup\u003e36\u003c/sup\u003e Recent studies have examined the potential use of bone marrow transplantation (BMT) to treat CSF1R-RD with highly conflicting results. For example, one study of 15 patients with a median follow-up of 26 months, found that BMT was associated with a significant worsening of cognitive symptoms.\u003csup\u003e37\u003c/sup\u003e In contrast, another recent study reported that BMT slowed cognitive decline in five patients, although one of these five subjects exhibited Montreal Cognitive Assessment (MoCA) and mini mental state exam (MMSE) scores of 0-1 out of 30 throughout the study.\u003c/p\u003e\n\u003cp\u003eMost recently, an elegant single nuclei sequencing (snSeq) study compared the brains of five ALSP subjects to twelve control subjects.\u003csup\u003e38\u003c/sup\u003e In addition to further demonstrating the loss of homeostatic microglia in ALSP brains, Du and colleagues detected a unique \u0026ldquo;ALSP-enriched microglia\u0026rdquo; signature that was virtually absent in controls. The authors further concluded that this unique cell cluster could either represent highly activated microglia or brain infiltrating \u0026ldquo;macrophages derived from peripheral monocytes\u0026rdquo;.\u003csup\u003e38\u003c/sup\u003e Interestingly, DEG analysis of this \u0026lsquo;ALSP microglia\u0026rsquo; cluster highlighted many genes that are also enriched in our current monocyte transcriptional signature, including \u003cem\u003eGPNMB, PLIN2, APOE, FTL, CD163, MITF, CHIT1, CR1, LGALS3, APOC1, ANXA2\u003c/em\u003e among others (Fig. 2d,f; Supplementary Table\u0026thinsp;1). We therefore sought to directly compare our monocyte and microglia signatures to those observed in human control and ALSP brains by Du et. al. Using our spatial sequencing dataset we generated module scores from the top 200 genes upregulated in monocyte or microglia engrafted brains. These module scores were then mapped onto control and ALSP patient UMAPs that were reconstructed from Du et. al.\u0026rsquo;s snSeq dataset (Fig. 5a-f). Consistent with our previously published comparisons between brain-engrafted iPSC-microglia and human resection-derived \u003cem\u003eex vivo\u003c/em\u003e microglia,\u003csup\u003e15\u003c/sup\u003e our microglial module score overlapped closely with homeostatic microglia identified by Du and colleagues (Fig. 5b,e). In contrast, our monocyte module score was greatly enriched within the unique cluster of \u0026lsquo;ALSP-microglia\u0026rsquo; identified by Du et. al., (Fig. 5f).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo further compare these datasets, we performed the inverse analysis, calculating \u0026lsquo;control-microglia\u0026rsquo; and \u0026lsquo;ALSP-microglia\u0026rsquo; module scores from Du et. al., and demonstrating a strong enrichment of these modules in our microglia or monocyte engrafted brains respectively (Fig. 5g). \u0026nbsp;In addition, we examined the log fold change in all significant myeloid DEGs detected by both studies. Regression analysis revealed a highly significant positive correlation (R\u003csup\u003e2\u003c/sup\u003e = 0.36, P\u0026lt;0.0001), further demonstrating the correspondence between Du and colleagues \u0026lsquo;ALSP microglia\u0026rsquo; and our current human monocyte engrafted samples. Lastly, we mapped the module scores calculated in Fig 5g onto our Visium spatial images, further demonstrating a strong enrichment of the ALSP-microglia module in our monocyte engrafted brains and a strong enrichment of the control homeostatic module in our iMG engrafted brains (Fig. 5i,j). Collectively, these analyses strongly suggest that the myeloid activation signature observed in our monocyte engrafted mice is closely related to a dysfunctional myeloid signature associated with a rapidly progressing and devastating human neurodegenerative disease. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMonocyte engraftment promotes cognitive dysfunction\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo determine the impact of the proinflammatory and hyperphagocytic profile of CNS-wide engrafted human monocytes on behavior, we quantified mouse exploratory behavior in the open field using Motion Sequencing (MoSeq), which uses machine learning to decompose movement into elemental motifs of action (e.g., runs, rears, turns) called \u0026ldquo;syllables\u0026rdquo;.\u003csup\u003e39\u003c/sup\u003e We quantified the expression of individual behavioral syllables expressed by transplanted mice at 5 months of age in an open-field arena over five consecutive days. We then subsequently performed three additional behavioral assays: rotarod, nesting, and marble burying (Fig 6a). In addition to revealing patterns of syllable usage, the MoSeq framework provides quantitative summary data r that revealed a striking difference in the position of mice transplanted with microglia versus monocytes as well as readily apparent difference in syllable usage (Fig. 6b). Consistent with pervasive behavioral differences between monocyte and microglia-transplanted mice, 78% of all syllables identified exhibited significant differences in usage in the open field (Fig 6c). Linear discriminant analysis (LDA) revealed a clear separation in the overall patten of syllable use between the transplanted mice that was stable over days (Fig. 6d). Visualization of the arena over 30 mins also revealed pronounced thigmotaxis within the monocyte engrafted mice as they\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003epaced along the arena walls (Fig. 6e). During the first minute of observation both monocyte and microglia mice exhibited typical murine behavior, exploring the border of the arena. However, over 30 minutes, monocyte engrafted mice continued to exhibit hyperactivity, showing no signs of habituation and leading to a significant increase in total distance traveled over 30 minutes (Fig. 6f,g).\u003c/p\u003e\n\u003cp\u003eSickness behavior in mice includes lethargy and reduced appetite and can be induced by proinflammatory stimuli such as lipopolysaccharide (LPS).\u003csup\u003e40\u003c/sup\u003e Similar to LPS, we found that monocyte engraftment leads to increased levels of multiple proinflammatory cytokines within both the brain and periphery (Fig. 4b, Extended Data Fig 3b). We therefore measured the weight of mice prior to sacrifice, detecting a significant reduction in the weight of monocyte engrafted mice (Fig. 6h). Despite this, monocyte and microglia engrafted mice exhibited equivalent motor function via accelerating Rotarod (Fig. 6i). However, in two other naturalistic tasks: marble burying and nest-building, monocyte engrafted mice exhibited greatly reduced activity (Fig. 6j,k). Low levels of nest building and burying activity have previously been shown to provide a sensitive measure of health and depression-like states and can be induced by\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003esystemic LPS treatment.\u003csup\u003e41,42\u003c/sup\u003e Therefore, to determine whether the observed behavioral impairments might be associated with reductions in weight, we performed a correlation analysis. This comparison revealed strong positive correlations between weight and both nest building (r=0.82, *P\u0026lt;0.013) and marble burying (r=0.83, *P\u0026lt;0.012), suggesting that the observed reductions in weight and naturalistic behavior in monocyte engrafted mice provide evidence of sickness behavior. Collectively, our behavioral analyses strongly indicate that widespread engraftment of human monocytes leads to significant neurobehavioral dysfunction.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eMicroglia arise early in development from HPCs within the yolk sac that give rise to primitive macrophages that migrate into the developing CNS.\u003csup\u003e11,12\u003c/sup\u003e\u0026nbsp; In contrast, blood monocytes and most other peripheral myeloid cells arise later in development and continue to be produced throughout life from definitive bone marrow-derived HSPCs.\u003csup\u003e11,12\u003c/sup\u003e Lineage tracing studies have further confirmed that the CNS receives little-to-no input from definitive HSPC-derived monocytes or macrophages in healthy mice.\u003csup\u003e10\u003c/sup\u003e Nevertheless, recent studies have coupled transplantation of definitive HSPCs, myeloid progenitors, or monocytes with microglia depleting treatments, to replace endogenous microglia with peripherally derived murine myeloid cells.\u003csup\u003e1-5\u003c/sup\u003e These initial studies have shown exciting evidence of reduced pathology and improved outcomes in mouse models of AD, LSD, and FTD. However, these along with other studies have also shown that engrafted MLCs remain transcriptionally and functionally distinct from microglia.\u003csup\u003e6-9\u003c/sup\u003e\u0026nbsp; As human and murine microglia also exhibit many species-dependent differences,\u003csup\u003e8,13-16\u003c/sup\u003e the potential impact of human MLCs on brain function remains unclear. It is therefore critically important to investigate the effects of human MLC engraftment in appropriate \u003cem\u003ein vivo\u003c/em\u003e systems before proceeding to clinical translation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe current study together with the accompanying report by Wang et. al., independently investigated the differences between human iPSC-microglia and peripherally derived monocytes following transplantation into the brains of adult microglia-deficient xenotolerant FIRE mice. Four months post-transplantation, despite similar numbers of engrafted cells and widespread brain distribution, we detected many important differences in the transcriptional and functional properties of brain engrafted monocytes versus microglia. Specifically, we find that human monocytes, but not microglia, led to a chronic proinflammatory state characterized by pronounced astrogliosis, demyelination, and neuronal and synaptic loss, culminating in severe behavioral impairments. Collectively, our data demonstrate that ontogeny plays a critical role in the long-term function of myeloid cells within the brain, suggesting that clinical development of approaches that promote the replacement of microglia with definitive myeloid cells instead of primitive-like iPSC-microglia warrants considerable caution.\u003c/p\u003e\n\u003cp\u003eA major strength of our findings is the remarkable similarities between our current results and the accompanying report by Wang and colleagues. Each study was conceived, pursued, and interpreted independently and yet comparisons between the two datasets reveal extremely strong concordance (Fig. 2h, Extended Data Fig. 1b,c). Importantly, the two studies also utilized complimentary approaches to examine the impact of myeloid cell ontogeny on brain health. Whereas Wang et.al., utilized epigenetic approaches to further understand the transcriptional regulation of definitive versus primitive myeloid ells, we utilized proteomic and behavioral approaches to further examine the downstream consequences of these differing ontogenies.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDespite these robust and consistent results, it is important to acknowledge some potential limitations of our study. To achieve brain wide engraftment of human monocytes in a model system we utilized immune deficient mice that lack, B, and NK cells. As a result, potential interactions between adaptive immunity and innate myeloid cells within the brain cannot be examined. The future development of more complex humanized models that employ matching definitive and primitive immune cell engraftment could provide a promising approach to address this shortcoming. Another potential caveat to our approach is that we examined the differences between complete brain wide monocyte versus microglia engraftment. Clinical approaches would be unlikely to fully replace endogenous microglia and thus understanding the potential interactions between engrafted monocytes and endogenous microglia will likely be informative. For example, can proinflammatory monocytes alter the activation state of endogenous microglia or vice versa? Future studies that examine co-transplantation of human microglia and monocytes could therefore be highly informative.\u003c/p\u003e\n\u003cp\u003eIn conclusion, the current study demonstrates the importance of human myeloid cell ontogeny on neuronal and glial health and behavioral function. Together with the accompanying study by Wang et. al., our findings suggest that CNS wide replacement of microglia with definitive human myeloid cells should be examined with considerable caution and that the use of stem cell derived microglia that better mimic the primitive ontogeny of microglia may provide a safer alternative. \u0026nbsp;\u0026nbsp;\u003c/p\u003e"},{"header":"METHODS","content":"\u003cp\u003e\u003cstrong\u003eAnimals\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll animal procedures were conducted in accordance with guidelines set forth by the National Institutes of Health and the University of California, Irvine Institutional Animal Care and Use Committee. Original CSF1R\u003csup\u003e\u0026Delta;FIRE/\u0026Delta;FIRE\u003c/sup\u003e mice were generated and previously characterized by Clare Pridans and David Hume.\u003csup\u003e43\u003c/sup\u003e FIRE mice were generated on a B6CBAF1/J background. Heterozygous FIRE mice were then crossed with wild-type C57BL/6J mice for four generations before backcrossing for four generations with xenotransplantation-compatible MITRG mice (Jackson Laboratories # 017711) which harbor deletions in Rag2 and IL2r\u0026gamma; and humanized CSF1, CSF2/Il3 and THPO. The resulting mice were maintained via breeding of heterozygous FIRE females (Fire\u003csup\u003e-/+\u003c/sup\u003e, Rag2\u003csup\u003e-/-\u003c/sup\u003e, il2rg\u003csup\u003e-/-\u003c/sup\u003e, hCSF1\u003csup\u003e+/+\u003c/sup\u003e, hCSF2/Il3\u003csup\u003e+/+\u003c/sup\u003e, hTHPO\u003csup\u003e+/+\u003c/sup\u003e) with homozygous (Fire\u003csup\u003e-/-\u003c/sup\u003e, Rag2\u003csup\u003e-/-\u003c/sup\u003e, il2rg\u003csup\u003e-/y\u003c/sup\u003e, hCSF1\u003csup\u003e+/+\u003c/sup\u003e, hCSF2/Il3\u003csup\u003e+/+\u003c/sup\u003e, hTHPO\u003csup\u003e+/+\u003c/sup\u003e) or heterozygous males (Fire\u003csup\u003e-/+\u003c/sup\u003e, Rag2\u003csup\u003e-/-\u003c/sup\u003e, il2rg\u003csup\u003e-/y\u003c/sup\u003e, hCSF1\u003csup\u003e+/+\u003c/sup\u003e, hCSF2/Il3\u003csup\u003e+/+\u003c/sup\u003e, hTHPO\u003csup\u003e+/+\u003c/sup\u003e). Genotyping of FIRE mice was performed using the following primers: \u0026nbsp;mCSF1R_FIRE_F (5\u0026rsquo;-GCGGTTGTAGGAAACCCTGA-3\u0026rsquo;), mCSF1R_WT_F (5\u0026rsquo;-GGTGCCAGCAATGTGTTTCC-3\u0026rsquo;), and mCSF1R_R: (5\u0026rsquo;-CACTCCTACCACTGGGCATC-3\u0026rsquo;). All mice were age- and sex-matched and group-housed, but single-housed post-transplantation, on a 12-h/12-h light/dark cycle with food and water \u003cem\u003ead libitum\u003c/em\u003e. Mice were housed with ambient temperature and humidity. Cages and bedding were changed every 1\u0026ndash;2 w. Behavioral analysis utilized hFIRE mice,\u003csup\u003e19\u003c/sup\u003e a highly similar xenotolerant model that lacks murine microglia but does not include humanized CSF2 \u003csup\u003e\u0026nbsp;\u003c/sup\u003eand THPO. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCell lines\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePeripheral Blood negatively sorted CD14\u003csup\u003e+\u003c/sup\u003e monocytes from four male subjects were obtained from Lonza (#4W-400; see details below). Four iPSC lines from healthy male subjects were generated and obtained from the Alzheimer\u0026rsquo;s Disease Research Core (ADRC) at University of California Irvine (see details in the table). Informed consent was received under an approved institutional IRB and Non-integrating Sendai virus (Life Technology, Cytotune 2.0) used to perform iPSC reprogramming. iPSCs clones were then confirmed to exhibit normal karyotype by Array Comparative Genomic Hybridization (ACGh by Cell Line Genomics), exhibit pluripotency via trilineage in vitro differentiation (Trilineage Differentiation Kit, Stem Cell Technologies), and sterility via MycoAlert (Lonza). \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.3965%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSample\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.3355%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCell Type\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.8042%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLine ID\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.02568%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSex\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.5088%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e(at donation)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRace\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSource\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.3965%;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.3355%;\"\u003e\n \u003cp\u003eCD14+ mono\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.8042%;\"\u003e\n \u003cp\u003e26198\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.02568%;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.5088%;\"\u003e\n \u003cp\u003e23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eCaucasian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eLonza\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.3965%;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.3355%;\"\u003e\n \u003cp\u003eCD14+ mono\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.8042%;\"\u003e\n \u003cp\u003e23617\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.02568%;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.5088%;\"\u003e\n \u003cp\u003e45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eCaucasian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eLonza\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.3965%;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.3355%;\"\u003e\n \u003cp\u003eCD14+ mono\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.8042%;\"\u003e\n \u003cp\u003e23623\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.02568%;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.5088%;\"\u003e\n \u003cp\u003e29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eUnknown\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eLonza\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.3965%;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.3355%;\"\u003e\n \u003cp\u003eCD14+ mono\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.8042%;\"\u003e\n \u003cp\u003e29631\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.02568%;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.5088%;\"\u003e\n \u003cp\u003e49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eCaucasian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eLonza\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.3965%;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.3355%;\"\u003e\n \u003cp\u003eiPSC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.8042%;\"\u003e\n \u003cp\u003eADRC 75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.02568%;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.5088%;\"\u003e\n \u003cp\u003e64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eCaucasian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eUCI ADRC\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.3965%;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.3355%;\"\u003e\n \u003cp\u003eiPSC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.8042%;\"\u003e\n \u003cp\u003eADRC 76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.02568%;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.5088%;\"\u003e\n \u003cp\u003e83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eCaucasian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eUCI ADRC\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.3965%;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.3355%;\"\u003e\n \u003cp\u003eiPSC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.8042%;\"\u003e\n \u003cp\u003eADRC 77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.02568%;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.5088%;\"\u003e\n \u003cp\u003e77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eCaucasian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eUCI ADRC\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.3965%;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.3355%;\"\u003e\n \u003cp\u003eiPSC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.8042%;\"\u003e\n \u003cp\u003eADRC 40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.02568%;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.5088%;\"\u003e\n \u003cp\u003e66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eAsian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 13.9647%;\"\u003e\n \u003cp\u003eUCI ADRC\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eMicroglia differentiation from iPSCs\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eiPSC-derived microglia (iMG) were differentiated using our previously published and highly replicated protocol.\u003csup\u003e44\u003c/sup\u003e To begin HPC differentiation, iPSCs were passaged onto 6-well matrigel-coated plates in TeSR-E8 at a density of 80 colonies of 100 cells each per 35 mm well. On day 0, cells were transferred to Medium A from the STEMdiff Hematopoietic Kit (05310; STEMCELL Technologies). On day 3, cells were exposed to Medium B and remained in Medium B for 7 additional days while small round HPCs began to lift off the colonies. On day 10, non-adherent CD43+ HPCs were collected by carefully removing medium and non-adherent cells with a serological pipette. At this point, HPCs were frozen in BamBanker freezing solution (Wako, NC9582225) for long-term storage. For iMG differentiation, HPCs were cultured on 6-well matrigel-coated plates in complete iMG medium (DMEM/F12 [11039-021; Thermo Fisher Scientific], 2\u0026times; insulin-transferrin-selenite [41400045; Thermo Fisher Scientific], 2\u0026times; B27 [17504044; Thermo Fisher Scientific], 0.5\u0026times; N2 [17502048; Thermo Fisher Scientific], 1\u0026times; Glutamax [35050-061; Thermo Fisher Scientific], 1\u0026times; non-essential amino acids [11140-050; Thermo Fisher Scientific], 400 mM monothioglycerol, and 5 mg/mL human insulin [I2643; Sigma-Aldrich] freshly supplemented with 100 ng/mL IL-34 [200-34; Peprotech], 50 ng/mL TGFb1 [100-21; Peprotech], and 25 ng/mL M-CSF [300-25; Peprotech]) for 28 days. Fresh complete iMG media was added every 24 h. iMG were frozen in BamBanker freezing solution (Wako, NC9582225) for long-term storage. Cells used for iMG transplantation were later thawed in complete iMG medium for 18\u0026ndash;24 h to recover before being resuspended at 62,500 cells/\u0026micro;l in 1\u0026times; Dulbecco\u0026apos;s Phosphate Buffered Saline (DPBS; low Ca2+, low Mg2+) for transplantation. A total of 500K cells (Monocyte or iMG) were injected per mouse.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAdult Intracranial Transplantation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll mouse surgeries and use were performed in strict accordance with approved National Institutes of Health (NIH) and institutional guidelines. Direct bilateral intracranial injections were performed as previously described.\u003csup\u003e15,45\u003c/sup\u003e Briefly, adult mice (2 m old) were anesthetized under continuous isoflurane and secured to a stereotaxic frame (Kopf), and local anesthetic (Lidocaine 2%; 17478-711-31, Medline) was applied to the head before exposing the skull. Using a 30-gauge needle affixed to a 10-mL Hamilton syringe, mice received 2 \u0026mu;L of cell suspension in sterile 1\u0026times; DPBS (14190144; Thermo Fisher Scientific) at 125,000 cells/\u0026mu;L at each injection site. Monocyte and iMG transplantations were conducted bilaterally into the lateral parietal association cortex and dorsal hippocampus at the following coordinates relative to bregma: anteroposterior, 2.06 mm; mediolateral, \u0026plusmn; 1.75 mm; dorsoventral, 1.75 mm (hippocampus) and 0.95 mm (cortex). Cells were injected at a rate of 62,500 cells/30 s with 4 min diffusion time in between injections. The needle was cleaned with consecutive washes of DPBS, 70% (vol/vol) ethanol, and DPBS in between hemispheres and animals. Animals were allowed to recover on heating pads before being placed in their home cages and received 2 mg/mL Acetaminophen (0904-7014-16; Major Pharmaceuticals) diluted in water for 5 d.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCollection and Isolation of Peripheral Blood Plasma\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBlood was collected into an EDTA-anticoagulant collection tube and immediately placed on ice. Tubes were then transferred to a pre-chilled centrifuge and spun at 1,600 rcf for 15 min at 4\u0026deg;C. Following centrifugation, plasma supernatant was gently collected and transferred to a 1.5 mL Eppendorf tube on ice. Eppendorf tubes were then transferred to a pre-chilled centrifuge and spun at 1,600 rcf for 15 min at 4\u0026deg;C. Following the second centrifugation, plasma supernatant was collected and aliquoted into 30 \u0026mu;L sub-aliquots and kept at -80\u0026deg;C for long-term storage.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIsolation of Soluble Protein Homogenates\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAnimals were perfused (20 mL per min) with 1X DPBS (4\u0026deg;C) for 3 min. Brains were surgically excised from each mouse and cut in half along the mid-sagittal plane. The left hemisphere was placed in 4% (wt/vol) paraformaldehyde (PFA) for 36 h at 4\u0026deg;C for subsequent immunohistochemistry (IHC) and spatial transcriptomics and the right hemisphere fresh-frozen on dry ice and stored at -80\u0026deg;C for biochemical analysis and proteomics. Fresh frozen hemispheres were crushed on liquid nitrogen using mortar and pestle and were split into 2 aliquots. One aliquot was used for Proteomics. The second aliquot was homogenized in solution of T-PER (78510, Thermo Scientific, Waltham, MA) and phosphatase and protease inhibitor mixtures (78426 and 78426, Thermo Scientific, MA). Homogenates were centrifuged at 16,000 rcf for 30 min at 4\u0026deg;C and supernatants were stored at \u0026minus;80\u0026deg;C for analysis. Protein concentration of soluble brain samples was determined using BCA Protein Assay Kit (Pierce) and 2 mg/mL of total protein per sample was run for ELISAs.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eQuantitative Biochemical Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAnalysis of soluble brain samples were conducted using commercially available ELISAs following manufacturer guidelines to measure Glial Fibrillary Acidic Protein (ABIN6574131; antibodies-online.com), Synaptophysin (ABIN6959799; antibodies-online.com), Postsynaptic density protein 95 (K15OQND; MSD), human XRCC5/Ku80 (MBS7608104; MyBioSource.com); human complement C4 (ab108824; Abcam); mouse IL-34 (M3400; R\u0026amp;D systems) and V-PLEX Proinflammatory Panel 1 Human Kit (K15049D; MSD). Analysis of peripheral blood plasma was also conducted using the V-PLEX Proinflammatory Panel 1 Human Kit (K15049D; MSD). Protein levels for each sample were then calculated via comparison to each assay-specific standard curve.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImmunohistochemistry\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAnimals were perfused with DPBS and isolated brain hemispheres were drop fixed in 4% (wt/vol) PFA for 36 h then cryoprotected in a 30% (wt/vol) sucrose at 4\u0026deg;C. Brains were sectioned coronally into 30-\u0026mu;m-thick slices on a freezing microtome (Leica, SM 2010R) and stored in a solution of 0.05% NaN3 (S2002; Sigma-Aldrich) in 1\u0026times; PBS (P44017-100TAB; Sigma-Aldrich) as free-floating slices. For staining, tissue was blocked for 1 h in 1\u0026times; PBS, 0.2% Triton X-100 (9002-93-1; Thermo Fisher Scientific), and 10% donkey serum (NC9624464; Thermo Fisher Scientific). Immediately following blocking, brain sections were placed in primary antibodies diluted in 1\u0026times; PBS and 1% donkey serum and incubated ON whilst shaking at 4\u0026deg;C. Samples were then incubated in conjugated secondary antibodies for 1 h followed by mounting on microscope slides. Sections were labeled with combinations of goat anti-IBA1 (1:300; ab5076; Abcam), rabbit anti-IBA1 (1:300; 019-19741; Wako), rabbit anti-Ku80 (1:200; ab80592; Abcam), mouse anti-Ku80 (1:250; MAS-12933; Invitrogen), rabbit anti-P2RY12 (1:500; HPA014518; Millipore Sigma), and chicken anti-GFAP (1:2000; ab4674; Abcam) overnight at 4\u0026deg;C followed by 1 h incubation at 23\u0026deg;C whilst shaking. After washing with 1\u0026times; PBS for 5 m three times, sections were incubated with highly cross-adsorbed AlexaFluor-conjugated secondary antibodies (1:400; ThermoFischer) for 1 h in the dark, then washed three times with 1\u0026times; PBS before mounting with Fluoromount-G (0100-01; SouthernBiotech).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImaging acquisition and processing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eImmunofluorescent sections were visualized and captured using an Olympus FX3000 confocal microscope using identical confocal and Z-stack settings for a given quantitative comparison. Half brain stitches to capture half-brains with high resolution were performed using Fluoview FV31S-DT software. For each quantitative analysis, three sections were imaged per region per animal using the \u0026ldquo;Surfaces\u0026rdquo; function in IMARIS software (Bitplane). All images were examined blinded using batch processing. \u0026nbsp;Numbers, area, volume, or intensity were calculated for each section, and these values were averaged to provide a single value per animal for subsequent statistical comparisons. For some representative images, brightness and contrast settings were slightly and equally adjusted across groups to reveal fine structures and morphology. Importantly, no changes were made to any images used for quantification.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSpatial transcriptomics and analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFixed half brains were cryoprotected in 30% sucrose and then sectioned in the coronal plane with a freezing sliding microtome. 10\u0026micro;m thick sections at the level of the dorsal hippocampus (Bregma AP -2.06) were collected and mounted on Superfrost plus glass microscope slides and stored at 4\u0026deg;C overnight to dry. Slide preparation was performed following the 10X Visium CytAssist Spatial protocol (Rev A). Each slide contained two capture areas, and each capture area was designed to hold three adjacent half brain coronal sections, allowing for parallel spatial profiling of multiple experimental conditions within a single assay. Sections were stained with hematoxylin and eosin (H\u0026amp;E) and imaged using brightfield microscopy to capture histological architecture. Spatial transcriptomic libraries were prepared using the 10x Genomics Human Transcriptome Probe Set v2, which uses probe hybridization and ligation-based capture (rather than poly-A enrichment), making it suitable for RNA derived from Paraformaldehyde fixed tissue. Although this probe set is designed to detect human transcripts, expression of genes such as GFAP, AQP4, and SNAP25\u0026mdash;typically associated with murine astrocytes and neurons\u0026mdash;was also observed. These signals were not computationally removed. Instead, it was assumed that myeloid-associated transcripts were derived from the engrafted human cells, as the host FIRE mouse lacks murine microglia. Library preparation, probe hybridization, and amplification were carried out by the UCI Genomics Core Facility, following the manufacturer\u0026rsquo;s protocols. Sequencing was performed on an Illumina NovaSeq 6000 platform using paired-end reads (28 bp for read 1 and 91 bp for read 2), achieving a target sequencing depth of 50,000\u0026ndash;100,000 reads per spot, consistent with recommended parameters for FFPE tissue. Each Visium spot represents pseudobulk gene expression from approximately 1 to 10 cells, depending on local tissue density and spot coverage.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTissue Demultiplexing and Brain Region Annotation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEach capture area included three distinct brain tissue sections, representing either biological replicates or different experimental conditions. To demultiplex the spatial data, manual segmentation was performed in Loupe Browser v8.0 or higher (10x Genomics). Using the aligned H\u0026amp;E-stained image, each tissue section was visually outlined based on its physical boundaries, and only barcodes falling entirely within each outlined area were extracted for downstream analysis. There was no spatial overlap between the three brain tissue sections within each capture area, ensuring that each barcode could be unambiguously assigned to a single tissue section. Spots located outside the H\u0026amp;E-detected tissue area were excluded from the analysis. To annotate anatomical brain structures, each tissue section was manually assigned to one of 13 predefined brain regions, including upper and lower cortical layers, the hippocampus, striatum, thalamus, amygdala, and white matter tracts (Fig 2a). These assignments were guided by visual alignment to reference coronal sections from the Allen Mouse Brain Atlas, using anatomical landmarks such as ventricles, white matter tracts, and cortical boundaries. The resulting barcode-to-region assignments were added to the Seurat object as metadata and were used in region-stratified analyses, including differential gene expression, module score comparisons, and spatial visualization.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Preprocessing and Quality Control\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSequencing reads were processed using Space Ranger v3.1.0 with the appropriate GRCh38 reference genome and FFPE Human Transcriptome Probe Set v2 annotation files (10x Genomics). Raw outputs\u0026mdash;including feature-barcode matrices, spatial image data, and tissue position files\u0026mdash;were imported into R (v4.4.2) using the Load10X_Spatial() function from Seurat v5.3.0. Given the fragmented nature of RNA in FFPE tissue, a rigorous quality control (QC) pipeline was applied to retain only high-quality barcodes corresponding to biologically informative tissue spots. Global distributions of key QC metrics were visualized using observed quantile plots, allowing empirical determination of filtering thresholds. The following filters were applied: Total UMI count (nCount_Spatial) \u0026gt; 800, Number of detected genes (nFeature_Spatial) \u0026gt; 300, Percentage of mitochondrial transcripts \u0026lt; 10%, Complexity score \u0026gt; 0.05. The complexity score was calculated as the ratio of unique genes to total UMI per spot: complexity = nFeature_Spatial / nCount_Spatial. This metric captures transcript diversity relative to sequencing depth and helped identify and exclude low-complexity (potentially damaged or ambient RNA\u0026ndash;dominated) barcodes. Barcodes outside the tissue area, as defined by H\u0026amp;E-based segmentation, were excluded during demultiplexing. Each spot was annotated with sample ID, experimental condition, slide ID, and capture area as metadata. Slide- and region-specific barcode prefixes were standardized to enable consistent downstream merging and integration.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNormalization, Integration, and Dimensionality Reduction\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEach sample was independently normalized using SCTransform, which stabilizes variance while adjusting for technical confounders such as sequencing depth and mitochondrial gene content. To account for batch effects due to differences in slide, capture area, and processing time, data were integrated using Harmony, which aligns datasets in principal component (PC) space. Slide ID and capture area were specified as covariates for batch correction, enabling alignment of biological signals across technical replicates while preserving spatial and anatomical structure. Dimensionality reduction was performed using Principal Component Analysis (PCA) followed by Uniform Manifold Approximation and Projection (UMAP) for two-dimensional visualization. Graph-based clustering was performed using Seurat\u0026rsquo;s FindNeighbors() and FindClusters() functions with resolutions ranging from 0.4 to 0.8. Clusters were assessed for biological granularity and stability across resolutions. A resolution of 0.6 was selected for downstream analyses, as it produced a non-redundant but sufficiently detailed set of clusters. This choice was informed by cluster tree analysis and best practices outlined in the Bioconductor OSCA workflow (https://bioconductor.org/books/3.12/OSCA/clustering.html). Spatial feature plots were generated using Seurat functions and maintained precise registration to H\u0026amp;E histological images, preserving anatomical context for downstream spatial analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDifferential Gene Expression Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo quantify transcriptional differences across conditions and regions, we applied two complementary frameworks\u0026mdash;spot-level testing and pseudobulk aggregation\u0026mdash;to retain spatial resolution while adding replicate-aware statistical control to limit false positives. Spot-level differential expression was performed on SCT-normalized data using Seurat\u0026rsquo;s FindMarkers() (Wilcoxon rank-sum), including batch covariates in the model. Significance was defined as Benjamini\u0026ndash;Hochberg\u0026ndash;adjusted P (FDR) \u0026lt; 0.05 and |log₂FC| \u0026gt; 1. This approach is sensitive to spatially localized or cell-type\u0026ndash;specific differences but treats spots as independent replicates, which can inflate false discoveries due to within-sample correlation\u0026mdash;a recognized limitation in single-cell/spatial analyses.\u003csup\u003e46,47\u003c/sup\u003e Pseudobulk differential expression was also performed to\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eincorporate biological replication and mitigate pseudo-replication, raw counts were\u0026nbsp;aggregated per sample\u0026nbsp;and analyzed with\u0026nbsp;edgeR v3.42.4 by contrasting Monocyte condition to iMG. Libraries were normalized using\u0026nbsp;TMM, and\u0026nbsp;GLMs were fit with\u0026nbsp;condition\u0026nbsp;and\u0026nbsp;batch covariates. Genes were called significant at\u0026nbsp;FDR \u0026lt; 0.05\u0026nbsp;and\u0026nbsp;|log₂FC| \u0026gt; 1. This replicate-aware framework improves control of false discovery and detects\u0026nbsp;global, consistent shifts\u0026nbsp;across samples. Combining both spot-level and pseudobulk frameworks balances\u0026nbsp;spatial sensitivity\u0026nbsp;(spot-level) with\u0026nbsp;statistical robustness\u0026nbsp;(pseudobulk). We prioritized\u0026nbsp;high-confidence DEGs\u0026nbsp;detected by\u0026nbsp;both\u0026nbsp;methods;\u0026nbsp;spot-only\u0026nbsp;DEGs were interpreted as localized/cell-type\u0026ndash;restricted signals representing spatially resolved transcriptional programs, whereas\u0026nbsp;pseudobulk-only\u0026nbsp;DEGs reflected broad but modest sample-level/condition specific shifts. This dual strategy follows best-practice recommendations for spatial/single-cell DE.\u003csup\u003e46\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eModule Score Calculation\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGene program activity per spatial transcriptomics spot was quantified using Seurat\u0026apos;s \u003cem\u003eAddModuleScore()\u003c/em\u003e function, applied to SCT-normalized expression values. We curated three transcriptional modules to capture distinct myeloid activation states relevant to the xenotransplant context: 1) Disease-Associated Microglia (DAM), 2) Homeostatic microglia (\u003cem\u003eP2RY12, P2RY13, CX3CR1, ACY3\u003c/em\u003e,\u003cem\u003e\u0026nbsp;MX1\u003c/em\u003e) and 3) Monocytes (\u003cem\u003eCD14, MSR1, LGALS1, GPNMB, ITGAX\u003c/em\u003e). Gene sets were manually curated based on prior literature and pathway relevance.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSpatial-seq versus RNA-seq Conmparisons\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGenes were plotted in a ratio\u0026ndash;ratio plot and points were colored according to significance and directionality thresholds. Overall concordance between datasets was quantified using the Pearson correlation coefficient. Venn plots comparing upregulated monocyte genes across the two studies compared genes that reached significance in both studies (Log2FC\u0026gt;1, FDR\u0026lt;0.05) via hypergeometric P.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eVisualization and Statistical Reporting\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eIntersection (Venn) analysis :\u0026nbsp;\u003c/em\u003eOverlap between spot-level and pseudobulk DEGs was summarized with VennDiagram with \u003cem\u003eggven\u003c/em\u003e reporting counts and percentages for each set and the intersection. Genes in the intersection were prioritized as high-confidence significant results for downstream summaries and figures.\u003cem\u003e\u0026nbsp;Volcano plots\u003c/em\u003e: Differential expression results were visualized with \u003cem\u003eEnhancedVolcano\u003c/em\u003e using the pseudobulk result as background, displaying log₂FC vs. \u0026ndash;log₁₀(FDR). The significant genes (intersection of the two methods) were colored coded by group (iMG vs. Monocyte), with selected marker genes labeled using \u003cem\u003eggrepel\u003c/em\u003e to avoid overlap.\u003cem\u003e\u0026nbsp;Violin plots with statistics:\u0026nbsp;\u003c/em\u003eExpression of selected genes and module scores was summarized with violin plots (ggplot2), showing median and interquartile ranges. Two-sided Wilcoxon rank-sum tests were performed using \u003cem\u003eggpubr::stat_compar\u003c/em\u003e\u003cem\u003ee_means\u003c/em\u003e, with Benjamini\u0026ndash;Hochberg correction applied when testing multiple genes/regions. Adjusted P values were displayed directly on the plots. Spatial maps genes and modules: Spatial distributions of single genes and module scores (DAM, IFN, iMG, Monocyte) were rendered with Seurat\u0026rsquo;s \u003cem\u003eSpatialFeaturePlot\u003c/em\u003e and \u003cem\u003eSpatialDimPlot\u003c/em\u003e, and, where helpful, customized using \u003cem\u003eSeuratExtend/scCustomize\u003c/em\u003e for consistent legends and image ordering across samples. All spatial plots were registered to H\u0026amp;E images.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eProteomics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTandem mass tag (TMT)\u0026ndash;labeled brain tissue lysates was processed as previously described.\u003csup\u003e48,49\u003c/sup\u003e Samples (100ug) were reduced and alkylated with 5\u0026thinsp;mM DTT and 10mM IAA in the dark for 30\u0026thinsp;min. The samples were then digested with 4ug of Lysyl endopeptidase (Wako) overnight. Samples were then diluted with 7X 100mM Tris with 4ug trypsin (ThermoFisher) and digested overnight at room temperature. The peptide solutions were acidified to a final concentration of 1% (vol/vol) formic acid (FA) and 0.1% (vol/vol) trifluoroacetic acid (TFA) then mixed 1:1 with 4% H3PO4 and desalted with a 10 mg Prime MXC plate (Water\u0026apos;s). Each MCX column was loaded directly after acidification. The wells were then washed with 500ul of L 100 mM NH4COOH / 2% formic acid and then with 500ul of methanol. Samples were eluted with 500ul of (50:50 ACN:MeOH + 5% NH4OH) and immediately dried by a SpeedVac. For TMT labeling, each brain peptide digest was resuspended in 75 \u0026micro;L of 100 mM triethylammonium bicarbonate (TEAB) buffer, and 5 mg of TMTpro reagents was dissolved in 200 \u0026micro;L ACN. Then 100 \u0026micro;g of peptide samples were aliquoted and resuspended in 100 \u0026micro;L TEAB buffer. After bringing the TMTpro reagents to room temperature and mixing with ACN, 41 \u0026micro;L of the TMT solution was added to each peptide solution and incubated for 1 h. The reaction was stopped with the addition of 8 \u0026micro;L of 5% hydroxylamine. The labeled samples were then combined, concentrated using a SpeedVac, and diluted with 0.1% TFA. After acidification, the peptides were desalted using a C18 Sep‐Pak column, washed, and eluted with 50% ACN. Each sample was assessed for complete labeling on a single-shot LC-MS/MS analysis prior high pH fractionation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHigh-pH Off-line Fractionation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDried samples were re-suspended in high pH loading buffer (0.07% vol/vol NH4OH, 0.045% vol/vol FA, 2% vol/vol ACN) and loaded onto a Water\u0026rsquo;s BEH 1.7 um 2.1mm by 150mm. A Thermo Vanquish was used to carry out the fractionation. Solvent A consisted of 0.0175% (vol/vol) NH4OH, 0.01125% (vol/vol) FA, and 2% (vol/vol) ACN; solvent B consisted of 0.0175% (vol/vol) NH4OH, 0.01125% (vol/vol) FA, and 90% (vol/vol) ACN. The sample elution was performed over a 25 min gradient with a flow rate of 0.6 mL/min. A total of 192 individual equal volume fractions were collected across the gradient and subsequently pooled by concatenation into 96 fractions and dried to completeness using a SpeedVac.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLiquid Chromatography\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEach sample was resuspended in 10 ul of loading buffer (0.1% FA) and 2 ul was analyzed by liquid chromatography coupled to tandem mass spectrometry. Peptide eluents were separated on Water\u0026rsquo;s CSH 150um x 15cm column packed with 1.7um resin. Buffer A was water with 0.1% (vol/vol) formic acid, and buffer B was 80% (vol/vol) acetonitrile in water with 0.1% (vol/vol) formic acid. Elution was performed over a 30 min gradient. The non-linear gradient was from 1% to 99% solvent B.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMass Spectrometry\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePeptides were monitored by Orbitrap Eclipse spectrometer (ThermoFisher) fitted with a high-field asymmetric waveform ion mobility spectrometry (FAIMS Pro) ion mobility source (ThermoFisher). Two compensation voltages (CV) of -45 and -65 were chosen for FAIMS. Each 1.5 second cycle consisted of one full scan (MS1) was performed with an m/z range of 410-1600 at 60,000 resolution, 100% AGC and 50 ms injection time and higher energy collision-induced dissociation (HCD) TMT scans collected at 30,000 resolutions with Turbo TMT on. The mass range was set to start at 120 m/z, isolation was set to 0.7 D and AGC was set to 250% with a max injection time of 86 ms.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDatabase Search\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe raw files were searched using FragPipe (version 19.1). The FragPipe pipeline relies on MSFragger (version 3.7)\u003csup\u003e50,51\u003c/sup\u003e for peptide identification and Philosopher (version 4.8.0)\u003csup\u003e52\u003c/sup\u003e for FDR filtering and downstream processing. The chimeric Uniprot protein database used here contains canonical isoforms for both Mouse (Mus musculus; accessed 02/2023) and Human (Homo sapiens, accessed 02/2019). The workflow used in FragPipe followed default TMT-16 plex parameters, used for both TMT-16 and TMT-18 experimental design. Briefly, precursor mass tolerance was -20 to 20 ppm, fragment mass tolerance of 20 ppm, mass calibration and parameter optimization were selected, and isotope error was set to -1/0/1/2/3. Enzyme specificity was set to strict-trypsin and up to two missing trypsin cleavages were allowed. Peptide length was allowed to range from 7 to 50 and peptide mass from either 200 to 5,000 Da. Variable modifications that were allowed in our search included: oxidation on methionine, N-terminal acetylation on protein, and N-terminal acetylation on peptide, with a maximum of 3 variable modifications per peptide. Peptide Spectral Matches were validated using Percolator.\u003csup\u003e53\u003c/sup\u003e The false discovery rate (FDR) threshold was set to 1% and protein and peptide abundances were quantified using Philosopher for downstream analysis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eProtein Quantitation and Quality Control\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe protein abundances are first normalized by scaling total protein signal within each channel for each specific sample to the maximum channel-specific total signal. We then used a tunable median polish approach, TAMPOR, to remove technical batch variance in the proteomic data, as previously described\u003csup\u003e54\u003c/sup\u003e TAMPOR is utilized to remove intra-batch and inter-batch variance while preserving meaningful biological variance in protein abundance values, normalizing to the median of selected intra-batch samples. This approach is robust to outliers and columns with up to 50% values missing. If a protein had more than 50% samples with missing values, it was removed from the matrix. No imputation of missing values was performed for any cohort. For the current data, TAMPOR leverages the median protein abundance from the pooled Global Internal Standard (GIS) TMT channels as the denominators in both factors to normalize sample-specific protein abundances across the 2 batches of samples.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eProteomic data processing and analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNormalized abundances were log₂-transformed prior to modeling. Differential abundance modeling: Differential protein abundance was assessed in R (v4.4.2) using limma (3.62.2) with empirical Bayes moderation. Condition Monocyte versus iMG was modeled as the primary effect; when multiple samples derived from the same donor were present, within-donor correlation was estimated with duplicateCorrelation(). Significance was defined as FDR (Benjamini\u0026ndash;Hochberg) \u0026lt; 0.05, with effect-size thresholds of |log₂FC| \u0026gt; 0.5 or \u0026gt; 1 as indicated. Quality control and visualization: Global data structure and sample comparability were evaluated by principal component analysis (PCA) and hierarchical clustering. Heatmaps were generated with pheatmap (1.0.13 using row-wise z-scores of log₂ abundances and annotated by condition. Volcano plots were produced with EnhancedVolcano (1.24.0) to display log fold change versus significance. \u003cem\u003eViolin plots:\u003c/em\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eFor each protein, log2-\u0026nbsp;transformed abundances were visualized as violin plots using ggplot2. Violin bodies show kernel density estimates; centered boxplots display the median and interquartile range. Proteins were ordered by the difference in mean expression between conditions (Monocyte \u0026minus; iMG). For each gene, differences between iMG and Monocyte were assessed with a two-sided, unpaired Wilcoxon rank-sum test (\u003cem\u003ewilcox.test via ggpubr::compare_means\u003c/em\u003e). The resulting p-values are shown in Figure 3c.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMoSeq\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e3D depth videos were recorded in a circular 17\u0026rdquo; diameter open-field arena (OFA) using a Microsoft Axure Kinect camera. Before recording, the mice were habituated in the examination room for 45 min. During the behavioral recording, the mice were freely moving in the arena for 30 minutes, then put back to its home cape. Between recording sessions, the area was cleaned according to the Moseq protocol. The recorded videos were analyzed using the MoSeq framework, available at http://moseq4all.org. MoSeq is an unsupervised machine learning framework that process and analyze animal spontaneous behavior into sub-second, repeated and modulated segments called \u0026ldquo;syllables\u0026rdquo;. The pipeline first processed the video such that the mouse was in the center of the video and the mouse was pointing right. The processed videos then went through dimensionality reduction and modeling to get the frame-by-frame behavioral labels. Syllable usages were aggregated by experimental group in the analysis. Syllable usage differences were identified using Kruskal-Wallis test and a post hoc Dunn\u0026rsquo;s two-sided test with permutation, using Bejamini-Hochberg FDR (\u0026alpha;=0.05).\u003csup\u003e55\u003c/sup\u003e Linear discriminant analysis (LDA) is a method that finds the linear combination of inputs to maximize group separation. To visualize the behavioral difference between the xenografted mice, LDA took syllable usages and experimental group-day pair as input to learn the low dimensional embedding. The position heatmap was computed by aggregating the mouse centroid positions and the heatmap represents the percentages of time. Distance traveled was computed by summing the between frame Euclidean distance between the mouse centroids in two consecutive frames within a given duration.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMarble Burying\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStandard polycarbonate mouse cages were fitted with filter-top covers and filled with 5mm of fresh, unscented bedding material. 15 glass marbles (3 rows, equally spaced, with 5 marbles per row) were placed at the surface of the bedding. Mice in their home cages were then transferred to the behavioral room to acclimate. After 45 min, each mouse was transferred to the behavioral cage. After 30 min, mice were transferred back to their individual home cages, and the number of buried marbles (\u0026gt;50% below the surface of the bedding) were scored by three blinded observers as previously described\u003csup\u003e56\u003c/sup\u003e whose scores were averaged per mouse.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNesting\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMice were transferred to a standard polycarbonate mouse cage with filter-top cover containing a single nestlet. After 24 hours, nest building was scored on a scale of 1-5 as previously described\u003csup\u003e57\u003c/sup\u003e by three blinded observers whose scores were averaged per mouse per day. Mice were then transferred to a new cage with a new nestlet and similarly scored a second time after 24 hours. Mice were then transferred to a new cage for the third time with a new nestlet and similarly scored after 24 hours. Final scores were calculated by averaging nesting score across three days per mouse.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistic Calculations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGraphPad Prism 9 was used to perform statistical tests and generate \u003cem\u003eP\u003c/em\u003e values. We used standard designation of \u003cem\u003eP\u003c/em\u003e values throughout the Figures (ns = not significant or \u003cem\u003ep\u003c/em\u003e\u0026ge;0.05; *\u003cem\u003ep\u003c/em\u003e\u0026lt;0.05; **\u003cem\u003ep\u003c/em\u003e\u0026lt;0.01; ***\u003cem\u003ep\u003c/em\u003e\u0026lt;0.001; ****p\u0026lt;0.0001). Values are represented as average mean \u0026plusmn; SEM. Details of number of replicates and the specific statistical test used are provided in the individual figure legends. Data represented for all antibodies calculated from an average of 2-3 matched coronal sections per animal. For quantification of NeuN density a Grubbs outlier test identified one statistical outlier which was removed. Because group distributions of NeuN counts deviated from normality, we utilized a non-parametric Kruskal\u0026ndash;Wallis test followed by Dunn\u0026rsquo;s multiple comparisons.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe spatial RNA-seq datasets are available through the GEO Super Series accession number (will be added prior to publication). Proteomic data will be made available via \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFurther information and requests for resources and reagents should be directed to and will be fulfilled by Mathew Blurton-Jones ([email protected]).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Drs. David Hume and Clare Pridans for generously providing the original immune intact FIRE-knockout mouse model. iPSC lines were generated by the UCI-ADRC iPS cell core funded by NIH P30AG066519. Figures 1a, 1b, 3a, and 6a were created with Biorender (https://www.biorender.com).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization, H.D., JP.C., M.B.-J.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMethodology, H.D., JP.C., G.E.-S., S.L., J.H., A.L.C., J.K.C., D.D., J.N., C.T., S.K.S., F.W., A.S., K.M., M.M., T.E.L., A.K.M., L.L., A.L.\u003c/p\u003e\n\u003cp\u003eQuantitative analysis, JP.C., H.D., A.L.\u003c/p\u003e\n\u003cp\u003eSpatial sequencing preparation and analysis, G.E.-S, J.K.C., J.H., A.L.\u003c/p\u003e\n\u003cp\u003eProteomic data collection and analysis, D.D., F.W, A.S., N.T.S., J.H., A.L.\u003c/p\u003e\n\u003cp\u003eWriting, JP.C., H.D. and M.B.-J.\u003c/p\u003e\n\u003cp\u003eWriting revisions, JP.C., H.D. M.B.-J., A.L., J.H., S.L., S.R.D., N.T.S.\u003c/p\u003e\n\u003cp\u003eSupervision, M.B.-J; N.T.S; S.R.D.\u003c/p\u003e\n\u003cp\u003eFunding acquisition, M.B.-J, JP.C., H.D., J.H., A.L.C., N.T.S., S.R.D., E.H.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported in part by the Larry L. Hillblom Foundation #2025-A-175-FEL, NIH T32 AG000096, and NIH T32 AG073088 (to JP.C.); NIH T32 AG000096 (to J.H.); Larry L. Hillblom Foundation #2024-A-023-FEL and BrightFocus Postdoctoral Fellowship Program in Alzheimer\u0026apos;s #A2025007F (to G.E.-S); CIRM EDUC4-12822 (to A.L.C.); NIH R01 AG061895 (to H.D.); NIH P30AG066519 (to E.H.); NIH R01AG073625 (to S.R.D); NIH U01AG061357 (N.T.S.); and NIH RF1NS139975, NIH R01AG082255, NIH U19AG069701, NIH P30AG066519, Alzheimer\u0026rsquo;s Association ADSF-25-1459986-C, and the Cure Alzheimer\u0026rsquo;s Fund Neuroimmune Consortium (to M.B.-J.). Generation of the ADRC iPSC lines were performed by the UCI-ADRC iPSC core funded by NIH P30AG16573.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eJP.C, H.D., J.H. and M.B.-J. are co-inventors on patent applications filed by the University of California Regents related to genetic modification of cells to confer resistance to CSF1R antagonists (WO2022/212897), transplantation of stem cell-derived microglia to treat leukodystrophies (WO2024/015933), and pathology-responsive recombinant cells (WO2023/212663). M.B.-J. is a co-founder and consultant for Savanna Biotherapeutics. No other disclosures were reported.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eYoo, Y., Neumayer, G., Shibuya, Y., Mader, M. M. \u0026amp; Wernig, M. A cell therapy approach to restore microglial Trem2 function in a mouse model of Alzheimer\u0026apos;s disease. \u003cem\u003eCell Stem Cell\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 1043-1053.e1046 (2023). https://doi.org/10.1016/j.stem.2023.07.006\u003c/li\u003e\n\u003cli\u003eP, C.\u003cem\u003e et al.\u003c/em\u003e CNS-wide repopulation by hematopoietic-derived microglia-like cells corrects progranulin deficiency in mice - PubMed. \u003cem\u003eNature communications\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e (07/05/2024). https://doi.org/10.1038/s41467-024-49908-4\u003c/li\u003e\n\u003cli\u003eShibuya, Y.\u003cem\u003e et al.\u003c/em\u003e Treatment of a genetic brain disease by CNS-wide microglia replacement. \u003cem\u003eSci Transl Med\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, eabl9945 (2022). https://doi.org/10.1126/scitranslmed.abl9945\u003c/li\u003e\n\u003cli\u003eFrosch, M.\u003cem\u003e et al.\u003c/em\u003e Microglia\u0026ndash;neuron crosstalk through Hex\u0026ndash;GM2\u0026ndash;MGL2 maintains brain homeostasis. \u003cem\u003eNature 2025\u003c/em\u003e (2025-08-06). https://doi.org/10.1038/s41586-025-09477-y\u003c/li\u003e\n\u003cli\u003eMader, M. M.-D.\u003cem\u003e et al.\u003c/em\u003e Therapeutic genetic restoration through allogeneic brain microglia replacement. \u003cem\u003eNature 2025\u003c/em\u003e (2025-08-06). https://doi.org/10.1038/s41586-025-09461-6\u003c/li\u003e\n\u003cli\u003eCronk, J. C.\u003cem\u003e et al.\u003c/em\u003e Peripherally derived macrophages can engraft the brain independent of irradiation and maintain an identity distinct from microglia. \u003cem\u003eJ Exp Med\u003c/em\u003e \u003cstrong\u003e215\u003c/strong\u003e, 1627-1647 (2018). https://doi.org/10.1084/jem.20180247\u003c/li\u003e\n\u003cli\u003eShemer, A.\u003cem\u003e et al.\u003c/em\u003e Engrafted parenchymal brain macrophages differ from microglia in transcriptome, chromatin landscape and response to challenge. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 5206 (2018). https://doi.org/10.1038/s41467-018-07548-5\u003c/li\u003e\n\u003cli\u003eJ, B.\u003cem\u003e et al.\u003c/em\u003e Monocytes can efficiently replace all brain macrophages and fetal liver monocytes can generate bona fide SALL1+ microglia - PubMed. \u003cem\u003eImmunity\u003c/em\u003e \u003cstrong\u003e58\u003c/strong\u003e (05/13/2025). https://doi.org/10.1016/j.immuni.2025.04.006\u003c/li\u003e\n\u003cli\u003eFC, B.\u003cem\u003e et al.\u003c/em\u003e A Combination of Ontogeny and CNS Environment Establishes Microglial Identity - PubMed. \u003cem\u003eNeuron\u003c/em\u003e \u003cstrong\u003e98\u003c/strong\u003e (06/27/2018). https://doi.org/10.1016/j.neuron.2018.05.014\u003c/li\u003e\n\u003cli\u003eA, S.\u003cem\u003e et al.\u003c/em\u003e Dual ontogeny of disease-associated microglia and disease inflammatory macrophages in aging and neurodegeneration - PubMed. \u003cem\u003eImmunity\u003c/em\u003e \u003cstrong\u003e55\u003c/strong\u003e (08/09/2022). https://doi.org/10.1016/j.immuni.2022.07.004\u003c/li\u003e\n\u003cli\u003eB, A., JL, B., C, K., KM, M. \u0026amp; FM, R. Infiltrating monocytes trigger EAE progression, but do not contribute to the resident microglia pool - PubMed. \u003cem\u003eNature neuroscience\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e (07/31/2011). https://doi.org/10.1038/nn.2887\u003c/li\u003e\n\u003cli\u003eMildner, A.\u003cem\u003e et al.\u003c/em\u003e Microglia in the adult brain arise from Ly-6ChiCCR2+ monocytes only under defined host conditions. \u003cem\u003eNature Neuroscience 2007 10:12\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e (2007-11-18). https://doi.org/10.1038/nn2015\u003c/li\u003e\n\u003cli\u003eD, G.\u003cem\u003e et al.\u003c/em\u003e An environment-dependent transcriptional network specifies human microglia identity - PubMed. \u003cem\u003eScience (New York, N.Y.)\u003c/em\u003e \u003cstrong\u003e356\u003c/strong\u003e (06/23/2017). https://doi.org/10.1126/science.aal3222\u003c/li\u003e\n\u003cli\u003eTF, G.\u003cem\u003e et al.\u003c/em\u003e Transcriptomic analysis of purified human cortical microglia reveals age-associated changes - PubMed. \u003cem\u003eNature neuroscience\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e (2017 Aug). https://doi.org/10.1038/nn.4597\u003c/li\u003e\n\u003cli\u003eHasselmann, J.\u003cem\u003e et al.\u003c/em\u003e Development of a Chimeric Model to Study and Manipulate Human Microglia In Vivo. \u003cem\u003eNeuron\u003c/em\u003e \u003cstrong\u003e103\u003c/strong\u003e, 1016-1033.e1010 (2019). https://doi.org/10.1016/j.neuron.2019.07.002\u003c/li\u003e\n\u003cli\u003eLund, H.\u003cem\u003e et al.\u003c/em\u003e Competitive repopulation of an empty microglial niche yields functionally distinct subsets of microglia-like cells. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 4845 (2018). https://doi.org/10.1038/s41467-018-07295-7\u003c/li\u003e\n\u003cli\u003eAbud, E. M.\u003cem\u003e et al.\u003c/em\u003e iPSC-Derived Human Microglia-like Cells to Study Neurological Diseases. \u003cem\u003eNeuron\u003c/em\u003e \u003cstrong\u003e94\u003c/strong\u003e, 278-293.e279 (2017). https://doi.org/10.1016/j.neuron.2017.03.042\u003c/li\u003e\n\u003cli\u003eCZ, H.\u003cem\u003e et al.\u003c/em\u003e Human microglia maturation is underpinned by specific gene regulatory networks - PubMed. \u003cem\u003eImmunity\u003c/em\u003e \u003cstrong\u003e56\u003c/strong\u003e (09/12/2023). https://doi.org/10.1016/j.immuni.2023.07.016\u003c/li\u003e\n\u003cli\u003eChadarevian, J. P.\u003cem\u003e et al.\u003c/em\u003e Therapeutic potential of human microglia transplantation in a chimeric model of CSF1R-related leukoencephalopathy. \u003cem\u003eNeuron\u003c/em\u003e (2024). https://doi.org/10.1016/j.neuron.2024.05.023\u003c/li\u003e\n\u003cli\u003eKeren-Shaul, H.\u003cem\u003e et al.\u003c/em\u003e A Unique Microglia Type Associated with Restricting Development of Alzheimer\u0026apos;s Disease. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e169\u003c/strong\u003e, 1276-1290.e1217 (2017). https://doi.org/10.1016/j.cell.2017.05.018\u003c/li\u003e\n\u003cli\u003eMcQuade, A.\u003cem\u003e et al.\u003c/em\u003e Gene expression and functional deficits underlie TREM2-knockout microglia responses in human models of Alzheimer\u0026apos;s disease. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 5370 (2020). https://doi.org/10.1038/s41467-020-19227-5\u003c/li\u003e\n\u003cli\u003eR, M.\u003cem\u003e et al.\u003c/em\u003e Xenografted human microglia display diverse transcriptomic states in response to Alzheimer\u0026apos;s disease-related amyloid-\u0026beta; pathology - PubMed. \u003cem\u003eNature neuroscience\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e (2024 May). https://doi.org/10.1038/s41593-024-01600-y\u003c/li\u003e\n\u003cli\u003eChadarevian, J.\u003cem\u003e et al.\u003c/em\u003e Harnessing human iPSC-microglia for CNS-wide delivery of disease-modifying proteins - PubMed. \u003cem\u003eCell stem cell\u003c/em\u003e (2025). https://doi.org/10.1016/j.stem.2025.03.009\u003c/li\u003e\n\u003cli\u003eD, K., T, U. \u0026amp; T, T. Functions and development of red pulp macrophages - PubMed. \u003cem\u003eMicrobiology and immunology\u003c/em\u003e \u003cstrong\u003e59\u003c/strong\u003e (2015 Feb). https://doi.org/10.1111/1348-0421.12228\u003c/li\u003e\n\u003cli\u003eL, D.-O.\u003cem\u003e et al.\u003c/em\u003e FEAST: A flow cytometry-based toolkit for interrogating microglial engulfment of synaptic and myelin proteins - PubMed. \u003cem\u003eNature communications\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e (09/27/2023). https://doi.org/10.1038/s41467-023-41448-7\u003c/li\u003e\n\u003cli\u003eJM, S.\u003cem\u003e et al.\u003c/em\u003e Lipofuscin-like autofluorescence within microglia and its impact on studying microglial engulfment - PubMed. \u003cem\u003eNature communications\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e (11/03/2023). https://doi.org/10.1038/s41467-023-42809-y\u003c/li\u003e\n\u003cli\u003eU, E. \u0026amp; AV, M. Defining microglial-synapse interactions - PubMed. \u003cem\u003eScience (New York, N.Y.)\u003c/em\u003e \u003cstrong\u003e381\u003c/strong\u003e (09/15/2023). https://doi.org/10.1126/science.adh7906\u003c/li\u003e\n\u003cli\u003eClaes, C.\u003cem\u003e et al.\u003c/em\u003e Plaque-associated human microglia accumulate lipid droplets in a chimeric model of Alzheimer\u0026apos;s disease. \u003cem\u003eMol Neurodegener\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 50 (2021). https://doi.org/10.1186/s13024-021-00473-0\u003c/li\u003e\n\u003cli\u003eMS, H.\u003cem\u003e et al.\u003c/em\u003e APOE4/4 is linked to damaging lipid droplets in Alzheimer\u0026apos;s disease microglia - PubMed. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e628\u003c/strong\u003e (2024 Apr). https://doi.org/10.1038/s41586-024-07185-7\u003c/li\u003e\n\u003cli\u003eA, L.\u003cem\u003e et al.\u003c/em\u003e Amelioration of Tau and ApoE4-linked glial lipid accumulation and neurodegeneration with an LXR agonist - PubMed. \u003cem\u003eNeuron\u003c/em\u003e \u003cstrong\u003e112\u003c/strong\u003e (02/07/2024). https://doi.org/10.1016/j.neuron.2023.10.023\u003c/li\u003e\n\u003cli\u003eMM, C.\u003cem\u003e et al.\u003c/em\u003e Steatosis drives monocyte-derived macrophage accumulation in human metabolic dysfunction-associated fatty liver disease - PubMed. \u003cem\u003eJHEP reports : innovation in hepatology\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e (08/11/2023). https://doi.org/10.1016/j.jhepr.2023.100877\u003c/li\u003e\n\u003cli\u003eNR, W. \u0026amp; KJ, M. Macrophage-derived foam cells in atherosclerosis: lessons from murine models and implications for therapy - PubMed. \u003cem\u003eCurrent drug targets\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e (2007 Dec). https://doi.org/10.2174/138945007783220597\u003c/li\u003e\n\u003cli\u003eM, v. E. \u0026amp; JMFG, A. The Unique Phenotype of Lipid-Laden Macrophages - PubMed. \u003cem\u003eInternational journal of molecular sciences\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e (04/14/2021). https://doi.org/10.3390/ijms22084039\u003c/li\u003e\n\u003cli\u003eWade, C., Runeckles, K., Chataway, J., Houlden, H. \u0026amp; Lynch, D. S. CSF1R-Related Disorder: Prevalence of CSF1R Variants and Their Clinical Significance in the UK Population. \u003cem\u003eNeurology: Genetics\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e (2024 Jul 16). https://doi.org/10.1212/NXG.0000000000200179\u003c/li\u003e\n\u003cli\u003eDulski, J.\u003cem\u003e et al.\u003c/em\u003e Global Presence and Penetrance of CSF1R-Related Disorder. \u003cem\u003eNeurology: Genetics\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e (2024 Sep 13). https://doi.org/10.1212/NXG.0000000000200187\u003c/li\u003e\n\u003cli\u003eKempthorne, L.\u003cem\u003e et al.\u003c/em\u003e Loss of homeostatic microglial phenotype in CSF1R-related Leukoencephalopathy. \u003cem\u003eActa Neuropathol Commun\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 72 (2020). https://doi.org/10.1186/s40478-020-00947-0\u003c/li\u003e\n\u003cli\u003eDulski, J.\u003cem\u003e et al.\u003c/em\u003e Hematopoietic Stem Cell Transplantation in. \u003cem\u003ePharmaceutics\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e (2022). https://doi.org/10.3390/pharmaceutics14122778\u003c/li\u003e\n\u003cli\u003eDu, S.\u003cem\u003e et al.\u003c/em\u003e Mutations in the human CSF1R gene impact microglia\u0026rsquo;s maintenance of brain white matter integrity. \u003cem\u003eNature Immunology 2025 26:7\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e (2025-06-26). https://doi.org/10.1038/s41590-025-02195-7\u003c/li\u003e\n\u003cli\u003eAB, W.\u003cem\u003e et al.\u003c/em\u003e Mapping Sub-Second Structure in Mouse Behavior - PubMed. \u003cem\u003eNeuron\u003c/em\u003e \u003cstrong\u003e88\u003c/strong\u003e (12/16/2015). https://doi.org/10.1016/j.neuron.2015.11.031\u003c/li\u003e\n\u003cli\u003eD, K., E, M. \u0026amp; N, I. Monitoring Pathogen-Induced Sickness in Mice and Rats - PubMed. \u003cem\u003eCurrent protocols in mouse biology\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e (06/19/2017). https://doi.org/10.1002/cpmo.27\u003c/li\u003e\n\u003cli\u003eTagawa, N., Tsuneoka, Y. \u0026amp; Funato, H. Nest-building behavior in laboratory mice: Multifunctional roles and neural mechanisms. \u003cem\u003eNeuroscience \u0026amp; Biobehavioral Reviews\u003c/em\u003e \u003cstrong\u003e178\u003c/strong\u003e (2025/11/01). https://doi.org/10.1016/j.neubiorev.2025.106377\u003c/li\u003e\n\u003cli\u003eJiang, X.\u003cem\u003e et al.\u003c/em\u003e Proanthocyanidin prevents lipopolysaccharide-induced depressive-like behavior in mice via neuroinflammatory pathway. \u003cem\u003eBrain Research Bulletin\u003c/em\u003e \u003cstrong\u003e135\u003c/strong\u003e (2017/10/01). https://doi.org/10.1016/j.brainresbull.2017.09.010\u003c/li\u003e\n\u003cli\u003eRojo, R.\u003cem\u003e et al.\u003c/em\u003e Deletion of a Csf1r enhancer selectively impacts CSF1R expression and development of tissue macrophage populations. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 3215 (2019). https://doi.org/10.1038/s41467-019-11053-8\u003c/li\u003e\n\u003cli\u003eMcQuade, A.\u003cem\u003e et al.\u003c/em\u003e Development and validation of a simplified method to generate human microglia from pluripotent stem cells. \u003cem\u003eMol Neurodegener\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 67 (2018). https://doi.org/10.1186/s13024-018-0297-x\u003c/li\u003e\n\u003cli\u003eChadarevian, J. P., Davtyan, H., Lombroso, S. I., Bennett, F. C. \u0026amp; Blurton-Jones, M. CRISPR generation of CSF1R-G795A human microglia for robust microglia replacement in a chimeric mouse model. \u003cem\u003eSTAR Protoc\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 102490 (2023). https://doi.org/10.1016/j.xpro.2023.102490\u003c/li\u003e\n\u003cli\u003eSquair, J. W.\u003cem\u003e et al.\u003c/em\u003e Confronting false discoveries in single-cell differential expression. \u003cem\u003eNature Communications 2021 12:1\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e (2021-09-28). https://doi.org/10.1038/s41467-021-25960-2\u003c/li\u003e\n\u003cli\u003eZimmerman, K. D.\u003cem\u003e et al.\u003c/em\u003e A practical solution to pseudoreplication bias in single-cell studies. \u003cem\u003eNature Communications 2021 12:1\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e (2021-02-02). https://doi.org/10.1038/s41467-021-21038-1\u003c/li\u003e\n\u003cli\u003eSeifar, F.\u003cem\u003e et al.\u003c/em\u003e Large-scale deep proteomic analysis in Alzheimer\u0026apos;s disease brain regions across race and ethnicity. \u003cem\u003eAlzheimers Dement\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 8878-8897 (2024). https://doi.org/10.1002/alz.14360\u003c/li\u003e\n\u003cli\u003eHigginbotham, L.\u003cem\u003e et al.\u003c/em\u003e Unbiased classification of the elderly human brain proteome resolves distinct clinical and pathophysiological subtypes of cognitive impairment. \u003cem\u003eNeurobiol Dis\u003c/em\u003e \u003cstrong\u003e186\u003c/strong\u003e, 106286 (2023). https://doi.org/10.1016/j.nbd.2023.106286\u003c/li\u003e\n\u003cli\u003eAT, K., FV, L., DM, A., D, M. \u0026amp; AI, N. MSFragger: ultrafast and comprehensive peptide identification in mass spectrometry-based proteomics - PubMed. \u003cem\u003eNature methods\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e (2017 May). https://doi.org/10.1038/nmeth.4256\u003c/li\u003e\n\u003cli\u003eF, Y.\u003cem\u003e et al.\u003c/em\u003e Identification of modified peptides using localization-aware open search - PubMed. \u003cem\u003eNature communications\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e (08/13/2020). https://doi.org/10.1038/s41467-020-17921-y\u003c/li\u003e\n\u003cli\u003eda Veiga Leprevost, F.\u003cem\u003e et al.\u003c/em\u003e Philosopher: a versatile toolkit for shotgun proteomics data analysis. \u003cem\u003eNature Methods 2020 17:9\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e (2020-07-15). https://doi.org/10.1038/s41592-020-0912-y\u003c/li\u003e\n\u003cli\u003eK\u0026auml;ll, L.\u003cem\u003e et al.\u003c/em\u003e Semi-supervised learning for peptide identification from shotgun proteomics datasets. \u003cem\u003eNature Methods 2007 4:11\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e (2007-10-21). https://doi.org/10.1038/nmeth1113\u003c/li\u003e\n\u003cli\u003eEB, D., NT, S. \u0026amp; ECB, J. Batch Correction and Harmonization of -Omics Datasets with a Tunable Median Polish of Ratio - PubMed. \u003cem\u003eFrontiers in systems biology\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e (2023). https://doi.org/10.3389/fsysb.2023.1092341\u003c/li\u003e\n\u003cli\u003eAB, W.\u003cem\u003e et al.\u003c/em\u003e Revealing the structure of pharmacobehavioral space through motion sequencing - PubMed. \u003cem\u003eNature neuroscience\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e (2020 Nov). https://doi.org/10.1038/s41593-020-00706-3\u003c/li\u003e\n\u003cli\u003eWitkin, J. M. Animal Models of Obsessive‐Compulsive Disorder. \u003cem\u003eCurrent Protocols in Neuroscience\u003c/em\u003e \u003cstrong\u003e45\u003c/strong\u003e (2008/10/01). https://doi.org/10.1002/0471142301.ns0930s45\u003c/li\u003e\n\u003cli\u003eDeacon, R. M. \u0026amp; Deacon, R. M. Assessing nest building in mice. \u003cem\u003eNature Protocols 2006 1:3\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e (2006-08-24). https://doi.org/10.1038/nprot.2006.170\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"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":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"ontogeny, microglia, monocyte, myeloid, macrophage, iPSC, pluripotent, xenotransplantation, hematopoietic stem and progenitor cells (HSPC), bone marrow transplantation (BMT)","lastPublishedDoi":"10.21203/rs.3.rs-7762816/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7762816/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Transplantation of peripherally derived myeloid cells in combination with microglia depletion is increasingly being examined as a potential therapy for various neurological disorders.1-9 However, primitive yolk sac derived microglia and definitive bone marrow derived monocytes and macrophages exhibit distinct developmental ontogenies.10-12 Important differences between murine and human myeloid cells further obscure the translational potential of this approach.8,13-16 To understand the impact of peripherally derived human myeloid cells on brain function and health we utilized a xenotolerant mouse model that lacks endogenous microglia. Using a combination of spatial sequencing, proteomics, histological, biochemical, and machine learning behavioral approaches we then examined the impact of brain wide engraftment of human induced pluripotent stem cell-derived microglia, which mimic the primitive ontogeny of microglia, in comparison to definitive bone marrow derived monocytes. Here we show that engraftment of human monocytes, but not microglia, promotes a chronic proinflammatory state associated with astrogliosis, demyelination, and synaptic and neuronal loss. These robust changes in myeloid cell activation state and brain homeostasis are in turn accompanied by significant impairments in cognitive function. Together with the accompanying study by Wang and colleagues, we demonstrate the critical impact of myeloid cell ontogeny on neuronal health and function, providing important implications for the rapidly progressing development and translation of brain-wide microglial replacement therapies.1-5","manuscriptTitle":"Brain wide engraftment of human monocytes induces neurodegeneration and cognitive dysfunction","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-10 07:49:47","doi":"10.21203/rs.3.rs-7762816/v1","editorialEvents":[],"status":"published","journal":{"display":false,"email":"[email protected]","identity":"nature","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"nature","sideBox":"Learn more about [Nature](http://www.nature.com/nature/)","snPcode":"","submissionUrl":"","title":"Nature","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"3113c869-94e4-4d20-89fe-6dbcb129b8ea","owner":[],"postedDate":"October 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":55676329,"name":"Biological sciences/Neuroscience/Neuroimmunology"},{"id":55676330,"name":"Biological sciences/Immunology/Innate immune cells/Monocytes and macrophages/Microglial cells"},{"id":55676331,"name":"Biological sciences/Stem cells/Haematopoietic stem cells"}],"tags":[],"updatedAt":"2026-04-16T16:06:59+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-10 07:49:47","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7762816","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7762816","identity":"rs-7762816","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}