{"paper_id":"25fa6536-8cc1-45cd-ba90-4f09a9535deb","body_text":"1\t\nEpigenetic control of microglial mitochondrial immunity by KAT7 drives \nAlzheimer’s disease pathogenesis \n \nYongqing Liu 1, Minghua Fan 2, Yingzhi Ye 1, Henry Yi Cheng 1, Shuying Sun 1,2,3, Zhaozhu \nQiu1,2,4* \n \n1Department of Physiology,  Pharmacology and Therapeutics, Johns Hopkins University \nSchool of Medicine, Baltimore, MD 21205, USA. \n \n2Solomon H. Snyder  Department of Neuroscience,  Johns Hopkins University School  of \nMedicine, Baltimore, MD 21205, USA. \n \n3Brain Science Institute, Johns Hopkins University School of Medicine, Baltimore, MD, USA. \n \n4Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, MD \n21205, USA. \n \n \n*Corresponding author. Email: zhaozhu@jhmi.edu \n \n \n \n \nABSTRACT \n \nMitochondrial DNA (mtDNA) -driven innate immune signaling sustains chronic \nneuroinflammation in neurological diseases such as Alzheimer’s disease (AD), yet how this \npathway is regulated in microglia remains poorly understood. Here, we identify the histone \nacetyltransferase KAT7 (HBO1) as a central epigenetic regulator that links chromatin \nremodeling to mitochondrial immune activation. KAT7 and its histone mark H3K14ac are \nelevated in microglia from 5×FAD mice and human AD brains. Integrative transcriptomic and \nepigenomic analyses reveal that KAT7 activates transcription of Cmpk2, a mitochondrial \nkinase essential for mtDNA synthesis. Loss of KAT7 reduces Cmpk2 expression, impairs \nmtDNA replication and release, and consequently suppresses cGAS-STING and NLRP3 \nsignaling. Importantly, both microglia-specific deletion and pharmacological inhibition  of \nKAT7 mitigate cytosolic mtDNA -induced neuroinflammation, decrease amyloid-β burden, \nrestore synaptic plasticity, and improve cogniti ve function in 5×FAD mice. Together, these \nfindings uncover an epigenetic -mitochondrial axis sustaining microglial pathogenicity and \nestablish KAT7 as a promising therapeutic target for AD.       \n \nINTRODUCTION \n \nAlzheimer’s disease (AD) is the most common cause of dementia, affecting tens of millions \nworldwide and imposing an ever -growing societal and economic burden 1. It is \nneuropathologically characterized by the accumulation of extracellular amyloid-β (Aβ) plaques \nand intracellular neurofibrillary tangles of hyperphosphorylated tau2. Yet, decades of research\t\ntargeting these hallmark protein s have yielded only modest clinical benefit , indicating that \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 2\t\nadditional mechanisms underlie AD pathogenesis3. Mounting evidence implica tes chronic \nneuroinflammation as a central disease mechanism 4-6. Genome-wide association studies have \nlinked many AD risk loci t o microglial pathways,  highlighting microglia, the brain-resident \nimmune cells, as key players in AD7-9. While microglia initially protect the brain by clearing \nmisfolded proteins, chronic activation drives their transition into a proinflammatory state \nmarked by excessive cytok ine production and loss of homeostatic functions 10. This \nmaladaptive state fuels a self-perpetuating cycle of inflammation and neurodegeneration. \n \nMitochondrial dysfunction is increasingly recognized as a critical driver of chronic microglial \ninflammation11-13. In both aging and AD brains,  microglia exhibit elevated levels of \nmitochondrial DNA (mtDNA) in the cytosol, where it acts as a potent damage-associated \nmolecular pattern14-16. Cytosolic mtDNA is sensed by cyclic GMP -AMP synthase (cGAS), \nwhich produces the second messenger cyclic GMP -AMP (cGAMP) to activate STING \n(stimulator of interferon genes) 12. Activation of this pathway promotes phosphorylation of \nTBK1 (TANK-binding kinase 1 ) and IRF3 (interferon regulatory factor 3), leading to type I \ninterferon induction and the release of proinflammatory cytokines 17. Persistent cGAS-STING \nactivation in microglia sustains a maladaptive inflammatory state that drives AD progression. \nNotably, genetic or pharmacological inhibition of this signaling pathway mitigates microglial \nactivation and alleviates AD-related pathology15,18, underscoring its pathogenic role. However, \nthe upstream mechanisms governing this cytosolic mtDNA-initiated inflammatory cascade in \nmicroglia remain largely unknown. \n \nEpigenetic regula tion provides a critical layer of control over gene expression, enabling \ntransient stimuli to be converted into long -lasting transcriptional programs19-21. Among these \nmechanisms, histone acetylation plays a pivotal role: acetylation of lysine residues on histone \ntails generally relaxes chromatin structure and facilitates transcription 22. This places histone \nacetyltransferases (HATs), also known as lysine acetyltransferases (KATs), at the core of \ntranscriptional reprogramming22; however, their contribution to microglial inflammation and \nAD pathogenesis remains poorly understood. Here, through gene expression profiling, we \nidentify KAT7 (al so known as HBO1), a member of the MYST family of HATs 23, as a key \nepigenetic regulator of neuroinflammation. By coupling histone acetylation to enhanced \nmitochondrial DNA synthesis and cGAS-STING activation in microglia, our work uncovers a \npreviously unrecognized epigenetic mechanism driving chronic neuroinflammation and \nhighlights KAT7 as a promising therapeutic target for AD. \n \nRESULTS \n \nExpression of the KAT7 complex is upregulated in micro glia from both 5×FAD mouse \nmodel and human AD brains \n \nTo determine whether epigenetic regulators of histone acetylation are altered during microglial \nactivation, we analyzed public RNA-seq data from primary mouse microglia24 and found that \nJade2, which encodes a scaffold subunit of the KAT7 complex, was selectively upregulated in \nresponse to lipopolysaccharide (LPS) stimulation (Fig. 1A). We next performed RNA-seq on \nLPS-treated BV2 cells, a mouse microglia -derived cell line, and also observed increased \nexpression of Jade2 in response to LPS (Fig. 1B), whereas expression of other HAT complexes \nremained largely unchanged. These findings were further validated by qPCR and western blot \nanalyses (Fig. 1C and Suppl. Fig. 1). Given that neuroinflammation is a hallmark of AD5, we \nexamined whether expression and activity  of the KAT7 complex are also elevated in this \ncontext. To this end, we isolated microglia from 6 -month-old 5×FAD mice, a well-\n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 3\t\ncharacterized AD model carrying five familial AD mutations 25, using CD11b microbeads  \n(Suppl. Fig. 2A-B). qPCR analysis revealed that the expression of Kat7 and two of its scaffold \nsubunits, Jade2 and Brpf2, was increased in microglia from 5×FAD mice compared with age-\nmatched wild-type (WT) littermate controls (Fig. 1D). RNAscope analysis further showed \nupregulation of  Kat7 and Jade2 specifically in microglia , but not in neurons or astrocytes  \n(Suppl. Fig. 2C -D). To extend these findings  to humans, we analyzed a publicly available \nRNA-seq dataset (GSE125050) from post-mortem superior frontal gyrus tissue of AD patients \nand healthy controls26. While expression of KAT7 and its other subunits  was unchanged, the \nscaffolds JADE2 and BRPF3 were upregulated in microglia from AD patients (Suppl. Fig. 2E). \nBecause suitable antibodies for KAT7 complex proteins were unavai lable, we leveraged the \nfact that KAT7 is the primary enzyme responsible for H3K14 acetylation (H3K14ac) in cells27-\n29 and performed immunostaining with an anti -H3K14ac antibody on brain sections.  \nConsistently, whereas microglial H3K14ac levels were low in WT controls, they wer e \nmarkedly elevated in 5×FAD mice ( Fig. 1E ), with no changes observed in neurons or \nastrocytes (Suppl. Fig. 3). Importantly, H3K14ac signals were also strongly increased in \nmicroglia from postmortem human AD brain s compared with healthy controls ( Fig. 1F). \nTogether, these results demonstrate that upregulation of the KAT7 complex and its histone \nacetylation mark accompanies microglial activation, implicating KAT7 in neuroinflammation \nand AD pathogenesis. \n \nKAT7 regulates LPS- and Aβ-induced inflammatory responses in microglia \n \nTo investigate the role of KAT7 in neuroinflammation, we generated Kat7-knockout (KO) \nBV2 microglial cells using CRIPSR-Cas9 (Fig. 2A) and employed LPS stimulation as a well-\nestablished model of inflammatory activation. Kat7 deletion markedly reduced LPS-induced \niNOS expression and secretion of the inflammatory cytokine IL-6 (Fig. 2B-D). Conversely, \noverexpression of WT KAT7 enhanced IL-6 production, whereas the catalytically inactive \nmutant (KAT7-E508Q) failed to do so  (Fig. 2E-F), indicating that the enzymatic activity of \nKAT7 is required for its pro-inflammatory function. We next examined the role of the scaffold \nsubunit JADE2 in neuroinflammation . Overexpression of JADE2 increased KAT7 protein \nlevels, suggesting that JADE2 stabilizes the KAT7 complex ( Suppl. Fig. 4A ). Moreover, \nJADE2 promoted the expression of pro -inflammatory factors in a manner dependent on its \ninteraction with KAT7 ( Suppl. Fig. 4B ). To validate these findings in primary cells, we \ncultured microglia from neonatal mice and transfected them with Kat7-specific siRNAs (Fig. \n2G). Consistently, Kat7 knockdown significantly attenuated LPS-induced IL-6 production (Fig. \n2H-J). Similarly, p harmacological inhibition of KAT7 with WM-3835, a potent small -\nmolecule inhibitor27, suppressed LPS-induced Nos2 and Il6 expression in a dose-dependent \nmanner (Fig. 2K). In addition to LPS, we further asked whether KAT7 regulates inflammation \ndriven by aggregated Aβ. Treatment of primary mouse microglia with oligomeric Aβ42 \nrobustly induced production  of the pro-inflammatory cytokines IL -6 and IL -1β, which was \nmarkedly attenuated by WM-3835 (Fig. 2L-O), supporting a role of KAT7 in mediating Aβ -\ninduced inflammatory responses. These results establish KAT7 as a critical driver of microglial \ninflammatory responses to both LPS and aggregated Aβ. \n \nCombined transcriptomic and epigenomic  profiling reveals CMPK2 as a critical \ntranscriptional target of KAT7 \n \nTo elucidate the molecular mechanisms by which KAT7 regulates neuroinflammation, we \nperformed RNA-seq to profile transcriptomic changes in WT and Kat7 KO BV2 cells with or \nwithout LPS treatment (Fig. 3A). Using an adjusted p-value < 0.05 and |log2 fold change| > 1 \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 4\t\nas cutoffs, we identified 1 ,074 upregulated genes in response to LPS  stimulation in WT cells \n(Fig. 3B). Deletion of Kat7 attenuated the induction of 110 of these genes (Fig. 3C-D). Gene \nontology (GO) analysis revealed that these\tKAT7-dependent genes were significantly enriched \nin pathways related to interferon signaling and inflammatory responses (Fig. 3D). In contrast, \nof the 276 genes downregulated by LPS in WT cells, only five were reversed by Kat7 deletion \n(Suppl. Fig. 5A ), indicating that KAT7 has minimal impact on LPS -induced transcriptional \nrepression. Given that KAT7 is primarily responsible for H3K14 acetylation, a histone mark \nassociated with transcriptional activation 29, we next performed CUT&Tag (Cleavage Under \nTargets and Tagmentation) analysis to map the genome -wide distribution of H3K14ac  (Fig. \n3E). Genomic distribution analysis revealed that most differential H3K14ac peaks were located \nin promoter regions : 38% (6,121 peaks) between WT_LPS and WT cells, and 40% (5,776 \npeaks) between KO_LPS and WT_LPS cells  (Suppl. Fig. 5B-C). Among the genes showing \nincreased H3K14ac enrichment at promoters in WT cells upon LPS stimulation, 244 exhibited \nreduced acetylation in Kat7 KO cells (Fig. 3E and Suppl. Fig. 5D). Integrative analysis of \nRNA-seq and CUT&Tag datasets identified 17 genes as potential direct transcriptional targets \nof KAT7 in res ponse to LPS stimulation ( Fig. 3F-G). The limited overlap likely reflects the \nstringent thresholds applied.  Validation by qPCR and quantitative chromatin \nimmunoprecipitation (qChIP) using KAT7 and H3K14ac antibodies (with H3K23ac antibody \nas a negative control) confirmed the profiling results  (Fig. 3H-I and Suppl. Fig. 5E-G). \nNotably, among the KAT7 -regulated genes, Cmpk2 (Cytidine/uridine monophosphate kinase \n2) was one of the most dramatically suppressed in Kat7 KO cells in response to LPS, \nhighlighting it as a key downstream target (Fig. 3H and Suppl. Fig. 5E). Importantly, Cmpk2 \nwas also robustly induced by oligomeric Aβ42 in primary mouse microglia, and this induction \nwas markedly attenuated by Kat7 knockdown (Fig. 3J). Collectively, these findings identify \nCmpk2 as a direct transcriptional target of KAT7 and suggest that it functions as a shared \ndownstream effector in microglial inflammatory responses to both LPS and Aβ stimulation. \n \nKAT7\t drives microglial innate immune signaling via CMPK2 -dependent mtDNA \nsynthesis \n \nCMPK2 encodes a mitochondrial nucleotide monophosphate kinase that acts as the critical \nrate-limiting enzyme , ensuring dNTP precursor availability and driving the dramatic \nupregulation of  mitochondrial DNA (m tDNA) synthesis during macrophage activation30. \nCMPK2-dependent mtDNA synthesis facilitates the release of  mtDNA into the cytoplasm, \nwhich subsequently activate the cGAS-STING signaling pathway  and the NLRP3 \ninflammasome30-32. Given that KAT7 regulates Cmpk2 expression during microglia activation, \nwe tested whether it modulates mtDNA replication , release, and downstream innate immune \nresponses. We first performed 5-ethynyl-2'-deoxyuridine (EdU) labeling, which preferentially \nincorporates into newly synthesized mtDNA and appears as bright cytoplasmic puncta in non-\nproliferating primary mouse microglia (Fig. 4A-B). Kat7 knockdown impaired LPS-induced \nmtDNA replication , an effect  rescued by WT CMPK2 but not by a catalytically inactive \nCMPK2 mutant (CMPK2-D330A) (Fig. 4A -B and Suppl. Fig. 6), indicating that KAT7 \npromotes CMPK2-dependent mtDNA synthesis in response to LPS. To assess mtDNA release, \nwe primed microglia with LPS followed by ATP treatment, a common method to induce \nmitochondrial stress and trigger innate immunity32. qPCR analysis of cytosolic fractions using \nprimers specific to the mitochondrial D-loop region revealed elevated mtDNA levels in control \nmicroglia (Fig. 4C), indicating increased release from damaged mitochondria. Notably, Kat7 \nknockdown reduced cytosolic mtDNA levels (Fig. 4 C). Consistently, it led to reduced \nphosphorylation of TBK1 (Ser172, p-TBK1) and IRF3 (Ser396, p-IRF3) (Fig. 4D-G), as well \nas diminished IL-1β production (Fig. 4H), indicating attenuated cGAS-STING activation and \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 5\t\nNLRP3 inflammasome signaling. Importantly, CMPK2 overexpression restored cytosolic \nmtDNA levels and innate immune signaling in Kat7 deficient cells, whereas the catalytically \ninactive mutant failed to do so (Fig. 4C-H). Together, these results demonstrate that KAT7 \norchestrates CMPK2 -dependent mtDNA synthesis and release, establishing it as a critical \nepigenetic regulator of innate immune pathways during microglial activation.  \n \nMicroglia-specific Kat7 deletion attenuates neuroinflammation, reduces  Aβ pathology , \nand improves cognition in 5×FAD mice \n \nGiven our findings of elevated microglial KAT7 complex expression in AD, and its critical \nrole in regulating neuroinflammation in vitro , we hypothesized that KAT7 promotes AD \npathogenesis in vivo by driving cytosolic mtDNA -induced innate immune signaling. To test \nthis, we generated Kat7 floxed mice using the efficient additions with single -strand DNA \ninserts CRISPR (Easi-CRISPR) method33. Exon 2 of Kat7 was flanked by loxP sites, and its \ndeletion induces a frameshift mutation resulting in functional KO (Suppl. Fig. 7). These mice \nwere crossed with Cx3cr1 -CreER mice34 to achieve microglia-specific knockout (Kat7 cKO) \nand subsequently bred with 5×FAD mice (Fig. 5A). Tamoxifen was administered at 2 months \nof age to avoid perturbing microglial developmen t, and biochemical, pathological, and \nbehavioral assessments were performed at 6 months. Microglia isolated from Kat7 cKO mice \nexhibited efficient Kat7 deletion (Fig. 5B). Notably, Cmpk2, which was markedly upregulated \nin 5×FAD microglia, was reduced by Kat7 deletion (Fig. 5C). Given KAT7’s role in mtDNA \nsynthesis and release, we assessed cytosolic mtDNA levels by qPCR analysis and found them \nelevated in 5×FAD microglia but significantly attenuated in Kat7 cKO; 5×FAD mice (Fig. 5D). \nCorrespondingly, phosphorylation of TBK1 and IRF3, key mediators of the cGAS -STING \npathway, was markedly reduced in Kat7 deficient microglia, along with decreased levels of the \nproinflammatory cytokines IL -1β and IL -6 (Fig. 5E -H).\t Further analysis revea led that \nmicroglial activation was suppressed in Kat7 cKO; 5×FAD mice, as evidenced by a reduced \nnumber of Iba1-positive microglia in the hippocampus (Fig. 5I) and diminished protein levels \nof Iba1 and phosphorylated p65 (p-p65) (Fig. 5J), a key NF-κB effector downstream of cGAS-\nSTING signaling35.\t These results demonstrate that microglial Kat7 deletion suppresses \nneuroinflammation in 5×FAD mice. \n \nGrowing evidence indicates that microglia-driven neuroinflammation promotes Aβ deposition \nand disrupts synaptic activity, ultimately contributing to cognitive decline 5,36,37. To determine \nwhether Kat7 deletion mitigates Aβ pathology, we performed Thioflavin S (TS) staining on \nbrain sections from 6-month-old 5×FAD mice. Microglia-specific Kat7 cKO mice exhibited a \npronounced reduction in Aβ plaque burden across multiple brain regions, including the cortex \nand hippocampus ( Fig. 6A-B). To evaluate synaptic plasticity, we performed field potential \nrecordings to measure  long-term potentiation (LTP) at Schaffer collateral -CA1 synapses in \nacute hippocampal slices. The LTP deficits characteristic of 5×FAD mice were significantly \nrescued by microglial Kat7 deletion (Fig. 6C-D). We next assessed hippocampus -dependent \nspatial learning and memory using the Morris water maze. In line with restored synaptic \nfunction, Kat7 cKO; 5×FAD mice demonstrated accelerated learning during training (Fig. 6E) \nand superior memory retention in the probe trial, spending more time in the target quadrant and \ncrossing the former platform location more frequently compared to 5×FAD controls (Fig. 6F-\nH). Notably, swimming speed was comparable among groups ( Fig. 6I ), ruling out motor \ndeficits as a confounding factor.  Collectively, these results demonstrate that microglial Kat7 \ndeletion reduces Aβ accumulation, restores synaptic function, and improves cognitive \nperformance in 5×FAD mice.  \n \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 6\t\nPharmacological inhibition of KAT7 mitigates neuroinflammation, reduces Aβ burden, \nand improves cognition in 5×FAD mice \n \nTo determine whether pharmacological inhibition of KAT7 confers protection against AD, we \nfirst examined the effects of the KAT7 inhibitor WM-3835 in primary mouse microglia treated \nwith oligomeric Aβ42. Consistent with our knockdown results (Fig. 3J and Fig. 4A-B), WM-\n3835 treatment suppressed Aβ -induced Cmpk2 upregulation and markedly reduced CMPK2-\ndependent mtDNA synthesis ( Fig. 7 A-B). Before evaluating KAT7 inhibition in vivo , we \nexamined whether Kat7 deletion impacts brain function beyond early development, given its \nneuronal expression  and essential roles in  neural stem cell differentiation and cortical \ndevelopment29. We generated excitatory neuron -specific Kat7 cKO mice using CamKII -Cre \n(Suppl. Fig. 8A-B), which induces deletion beginning at 2-3 weeks of age38. Unlike mice with \nembryonic Kat7 deletion29, these cKO mice  were viable, healthy, and fertile. \nElectrophysiological recordings from hippocampal CA1 pyramidal neurons in acute slices \nrevealed normal miniature excitatory postsynaptic current (mEPSC) amplitude and frequency, \nas well as intact LTP at Schaffer collateral -CA1 synapses (Suppl. Fig. 8C-D), indicating \npreserved synaptic transmission and plasticity. These results suggest that loss of KAT7 is well \ntolerated in mature neurons. We then evaluated the therapeutic potential of WM-3835 in vivo. \nFive-month-old 5×FAD mice received intracerebroventricular (ICV) infusion of WM-3835 for \nfour weeks via an osmotic pump ( Fig. 7C). Post-treatment analysis confirmed robust target \nengagement, evidenced by a pronounced reduction in H3K14ac levels across the brain ( Fig. \n7D-E). Notably, WM-3835 significantly decreased microgliosis ( Fig. 7F), lowered cytosolic \nmtDNA levels (Fig. 7G), and attenuated cGAS-STING pathway activation in microglia (Fig. \n7H). Consistent with these effects, WM-3835 treatment also markedly reduced Aβ plaque \nburden in the hippocampus and cortex (Fig. 7I). Behaviorally, WM-3835-treated 5×FAD mice \nexhibited improved spatial learning and memory in the Morris water maze, with faster \nacquisition during training and superior performance in the probe trial ( Fig. 7J -L). These \ncognitive benefits closely mirrored those observed in microglia -specific Kat7 cKO mice. \nTogether, our results demonstrate that pharmacological inhibition of KAT7 mitigates \nmicroglial activation, reduces Aβ pathology, and improves cognitive function in 5×FAD mice, \nproviding preclinical evidence for WM-3835 as a potential therapeutic strategy for AD. \n \nDISCUSSION \n \nCytosolic mtDNA has emerged as a potent driver of neuroinflammation through activation of \ninnate immune signaling . Here, we identify the histone acetyltransferase KAT7 as a critical \nupstream regulator of this mtDNA -initiated inflammatory cascade in microglia  (Fig. 8 ). \nMechanistically, KAT7 promotes H3K14 acetylation at the Cmpk2 promoter during microglial \nactivation, thereby enhancing Cmpk2 expression, increasing mtDNA synthesis and release, and \nfueling downstream inflammatory responses. Multiple lines of evidence ind icate that this \npathway is active in AD. First, the expression of KAT7 complex and its histone mark H3K14ac \nare elevated in microglia from 5×FAD mice and human AD brains. Second, Aβ oligomers \ninduce Cmpk2 expression and mtDNA replication in primary microglia, processes that depend \nin part on KAT7 expression or activity. Third, Cmpk2 expression and cytosolic mtDNA levels \nare elevated in microglia from 5×FAD mice, both of which are reduced upon Kat7 deletion. \nFinally, both microglia-specific Kat7 deletion and pharmacological inhibition suppress innate \nimmune signaling and neuroinflammation, leading to reduced Aβ burden and improved \ncognitive function in 5×FAD mice. Our findings thus reveal an epigenetic-mitochondrial axis \nthat mechanistically link s Aβ deposition to chronic immune activation. Disrupting this self-\n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 7\t\nperpetuating loop offers a potential therapeutic strategy for AD and related neurodegenerative \ndisorders.  \n \nMicroglial activation requires extensive  metabolic reprogramming to meet heightened \nbioenergetic and biosynthetic demands 39,40. KAT7-meidated upregulation of CMPK2, a key \nenzyme for mtDNA replication, may help fulfill these needs. However, while essential for \nsustaining microglial activation, excessive mtDNA replication renders mitochondria \nvulnerable to stress -induced mtDNA release, thereby amplifying innate immune sign aling. \nNotably, a similar CMPK2 -mtDNA pathway drives cGAS -STING and NLRP3 activation in \nmacrophages during inflammation 30,32,41, underscoring its conserved role in innate immunity \nacross myeloid lineages. Future studies will be required to directly define the role of CMPK2  \nand its downstream mtDNA replication in neuroinflammation and AD. Of note, a recent study \nshowed that CMPK2 is also upregulated in microglia during ischemic stroke, where it promotes \nneuroinflammation and brain injury42. Therefore, the KAT7-CMPK2 pathway may represent a \ngeneral mechanism by which microg lia couple metabolic reprogramming to innate immune \nactivation across diverse chronic inflammatory conditions.  \n \nHistone acetylation in neurons has long been associated with learning and memory 43-45, and \nseveral altered histone acetylation marks have been reported in AD mouse models and \npatients46-48. However, the functional relevance of individual histone modifications in AD \nremains poorly defined. Here, we show that KAT7-dependent H3K14ac is low in microglia \nfrom WT mice and healthy controls  but markedly and specifically elevated in AD brains . \nFunctional studies further reveal a microglia -specific pathogenic role for  KAT7 in 5×FAD \nmice. These findings  highlight the importance of examining histone acetylation  and other \nepigenetic modifications  in a cell type -specific manner. Notably, KAT7 has also been \nimplicated in ageing and cellular senescence,\twhere it enhances transcription of the cyclin -\ndependent kinase inhibitor p15INK4b to promote cell cycle arrest49. Downregulation of Kat7 in \naged mice reduces hepatocyte senescence, alleviates liver inflammation, and extends lifespan49. \nGiven its beneficial effects in both AD and aging  contexts, KAT7 inhibition may represent a \npromising strategy to mitigate neurodegeneration and promote brain resilience across the \nlifespan. \n \nOur data establish CMPK2 as a key target of KAT7, but additional downstre am genes , \nincluding several interferon-responsive genes , are also likely to contribute to  chronic \nneuroinflammation. While KAT7 is best known for catalyzing H3K14ac, it also mediates other \nhistone acylations, such as propionylation and crotonylation50, and can modify both alternative \nhistone sites and non-histone substrates 23,51. These \t activities may further diversify it s \nregulatory functions and shape the transcriptional landscape of activated microglia. Regardless, \npharmacological inhibition of KAT7 is expected to suppress all of these activities, providing a \nunified strategy to blunt its proinflammatory effects in AD.  While the 5×FAD model used in \nthis study recapitulates robust Aβ pathology, it does not develop tau pathology, a defining \nfeature of human AD. Future studies employing tauopathy or mixed pathology models will \ntherefore be essential to fully delineate the role of KAT7 in AD pathogenesis. Nevertheless, \nour findings identify the KAT7-CMPK2 axis as a critical regulator of microglial mitochondrial \nimmunity, linking epigenetic control to neuroinflammation and AD progression. These insights \nhighlight the epigenetic-mitochondrial axis as a promising therapeutic target not only for AD \nbut also for other neurological diseases characterized by chronic neuroinflammation. \n \n \n \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 8\t\nMATERIALS AND METHODS \n \nHuman brain samples \nFormalin-fixed paraffin-embedded (FFPE) human postmortem brain samples from AD patients \nand controls (aged -matched and died of certain cause unrelated to dementia) were obtained \nfrom Johns Hopkins Brain Resource Center. Subject demographics were listed in Tables S1. \nThe study using patient samples/data was approved by the Johns Hopkins University School \nof Medicine Office of Human Subjects Research Institutional Review Boards. \n \nMice \nAll procedures related to animal care and treatment were approved by the Jo hns Hopkins \nUniversity Animal Care and Use Committee and met the guidelines of the National Institute of \nHealth Guide for the Care and Use of Laboratory Animals. All animals were group housed in \na standard 12 -hour light/dark cycle with ad libitum access to  food and water. The following \nmouse lines were used for the experiments: C57BL/6J (Jackson Laboratory, 000664), 5×FAD \n(Jackson Laboratory, 008730), Cx3cr1 -CreER (Jackson Laboratory, 021160) , Camk2a -Cre \n(Jackson Laboratory, 005359). Kat7 floxed mice were generated at Transgenic Core of Johns \nHopkins University. Male mice were used for all experiments unless otherwise noted. Female \nmice were also used  for the behavioral test s. Mice with Kat7 specific knockout in microglia \nwere induced by tamoxife n (S1238, Selleck). Briefly, tamoxifen was dissolved in corn oil \n(C8267, Sigma) to a final concentration of 20 mg/ml. 2 -month-old mice were administered \ntamoxifen via intraperitoneal injection at 100 mg/kg for five consecutive days.    \n \nGeneration of Kat7 floxed mice \nKat7 floxed mice were generated at Transgenic Core of Johns Hopkins University using the \nEasi-CRISPR method, as previously described 33. Two single -guide RNAs (sgRNAs) were \ndesigned by http://crispor.tefor.net/. The sequences were as follows: sgRNA #1 (reverse strand), \nAAGTACCAAGTTCCAACATAAGG; and sgRNA #2 (forward strand), \nGATACTGCTCCTGAGCTTGATGG. Two crRNAs  containing each sgRNA and ssDNA \ndonor containing the homology arms and floxed exon sequences were custom synthesized from \nIDT company. The annealed crRNA and tracrRNA (IDT) were diluted in microinjection buffer \n(0.25 mM EDTA and 10 mM Tris-HCl, pH 7.4) and mixed with Cas9 protein (30 ng/µl; IDT) \nto obtain ctRNP complexes. One -cell embryos of C57BL/6J mice were microinjected with a \nmixture of floxing ssDNA donors and two ctRNP complexes and were transferred into the \noviducts of pseudopregnant ICR females (C harles River Laboratories). Successful insertions \nof two LoxP sites were detected by PCR genotyping of mouse tails and confirmed by Sanger \nsequencing. The primers used are provided in Table S2. \n \nPrimary microglial cell preparation and stimulation \nPrimary microglial cells were prepared from neonatal mice (day 0 -2). Briefly, brain tissues \nwere quickly removed, and the meninges were carefully stripped in ice-cold HBSS. The cortex \nand hippocampi were then digested with 0.25% trypsin (Quality Biological) at 37 ℃ for 15 \nmin and gently pipetted to generate single cells with DMEM containing 10% heat -inactivated \nfetal bovine serum (FBS; Avantor ) and 1% penicillin/streptomycin (P/S; Quality Biological), \nfollowed by plating on poly-D-lysine-coated T75 flasks. After 11-14 days, primary microglia \nwere separated from the mixed glial cult ure using a shake -off method ( 90 rpm for 2 hours). \nThe collected microglia were seeded in the poly -D-lysine-coated plates and its purity was \nconfirmed by Iba1 immunostaining. \n \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 9\t\nFor LPS stimulation, microglial cells were cultured in basic DMEM without FBS and treated \nwith LPS (0.2 µg/ml, Sigma) for 6 h. For the assessment of mtDNA release, microglia were \nprimed with LPS (0.2 µg/ml, Sigma) for 6 h, followed by ATP (2 mM, Sigma) treatment for \n30 min. For oligomeric Aβ 42 stimulation, microglial cells were cultured in basic DMEM \nwithout FBS and treated with oligomeric Aβ42 (1 µM, rPeptide) for 6 h or 24 h.  \n \nIsolation of microglia from adult mouse brain \n6-month-old mice were anesthetized with isofl urane and perfused transcardially with cold \nsaline. Brain tissue was freshly harvested, cut into small pieces, and digested with collagenase \n(Type IV, 5 mg/ml, Sigma) and DNase I (50 µg/ml, Sigma) for 1 hour at 37 ℃ with 250 rpm. \nThe digested brain tissues were transferred to a 15 ml Dounce homogenizer and homogenized \ngently on ice. Brain tissue homogenates were suspended in HBSS, filtered with cell strainers \n(70 µm), and centrifuged at 500g for 5 min (4 °C) to collect the cell pellets. Then, 90% Percoll \nsolution was prepared using absolute Percoll (Cytiva) and 10× HBSS (9:1 , v/v), and further \ndiluted (v/v) to 70, 37, and 30% with 1× HBSS. Cell pellets were suspended in a 37% Percoll \nsolution. Microglia were isolated by density gra dient centrifugation. Density gradient was \nadded into 15 ml tubes, by layers of Percoll solution from bottom to top containing: 70%, 37%, \nand 30% Percoll solution and HBSS. Centrifugation was carried out in a horizontal centrifuge \nat 2000g for 30 min (4 °C). Microglia were converged on the interphase between the 37% and \n70% Percoll solution. Isolated microglia were washed with 10× volumes of PBS and \ncentrifuged at 500g for 5 min (4  °C). Microglia was further purified by CD11b MicroBeads \n(Miltenyi Biotec, 130-093-634) according to the manufacturer’s protocol.  \n \nBV2 cell culture \nThe mouse microglial BV2 cell line was a gift from Dr. Tony Wyss -Coray’s laboratory at \nStanford University52. Cells were cultured in DMEM supplemented with 10% FBS and 1% \npenicillin/streptomycin and maintained in an incubator at 37 °C with 5% CO2. Adherent cells \nwere split using 1× TrypLE (Gibco).  \n \nKAT7-KO BV2 cells were generated using CRISPR -Cas9 method. Guide RNA \n(GACTCGGGCAGATCGGCGCG) targeting mouse Kat7 was cloned in to LentiCRISPR-v2-\nPuro (Addgene, #98290). The primers used to design the single -guide RNA (sgRNA) targets \nwere (5’ to 3’) Kat7 forward CACCGGACTCGGGCAGATCGGCGCG and Kat7 reverse \nAAACCGCGCCGATCTGCCCGAGTCC. Lentiviral particles containing Kat7 sgRNA were \npackaged using the 3rd generation lentivirus system and used to infect BV2 cells. One day after \ninfection, the medium was changed to fresh DMEM containing 10% FBS and 1% P/S. Cells \nwere then treated with puromycin (4 µg/ml) for 5 days to select for successfully transduced \ncells. Single clon es were obtained using limiting dilution and were analyzed by western \nblotting and Sanger sequencing to confirm KAT7 deletion. A scrambled gRNA control was \nalso used as negative control (5’-GCGCCAAACGTGCCCTGACG-3’). \n \nFor the overexpression in BV2 cells, the GFP in the lentivirus vector pLenti-EF1a-GFP-P2A-\nPuro (a gift from Dr. Shuying Sun lab at Johns Hopkins University) was replaced by mouse \nCMPK2, CMPK2-D330A, human KAT7, KAT7 -E508Q and JADE2 DNA fragments at the \nAgeI and BamHI site using NEBuilder HiFi D NA Assembly Cloning Kit (New England \nBiolabs, #E5520). Lentiviral particles were packaged using the 3rd generation lentivirus \nsystem and used to infect BV2 microglia. Cells were then treated with puromycin (4 µg/ml) to \nselect the transduced cells. \n \nsiRNA transfection  \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 10\t\nsiRNAs were purchased from Dharmacon and transfected into primary microglia with the final \nconcentration of 40 nM using Lipofectamine RNAiMAX  (Invitrogen) according to the \nmanufacturer’s instructions. The sequences of siRNA used in this study are as follows: \nsiControl sense, UGGUUUACAUGUCGACUAA; mouse Kat7 siRNA#1 sense, \nGAACCGAAGAUUCCGAUUU; siRNA#2 sense, UGUUUGAAGUAGACGGCAA.  \n \nEnzyme-linked immunosorbent assay (ELISA) \nThe collected cultured medium of BV2 cells or primary microglia was centrifuged at 500 g for \n5 min (4°C) and the supernatant was processed for analysis with mouse IL -6 (431301, \nBioLegend) and IL -1β (432601, BioLegend) ELISA kits,  according to the manufacturer’s \ninstructions. For cortical IL-6 and IL -1β detection , the cortex tissues were removed from \nindicated mice and rapidly immersed in RIPA buffer (Sigma) containing  protease inhibitor \ncocktails (Roche). Total IL -6 and IL-1β protein levels were measured by the ELISA kit and \nnormalized first to total protein level quantified by the Pierce BCA Protein Assay Kit (Thermo) \nand then to the floxed (WT) group. \n \nWestern blotting \nProteins were isolated from cultured cells or brain tissues with RIPA buffer (Sigma) containing \nprotease inhibitor  cocktails (Roche). Samples were separated on Novex Tris -Glycine Mini \nProtein Gels (4 to 20%, Invitrogen) and transferred to nitrocellulose membranes (Bio -Rad), \nwhich were incubated with appropriate  antibodies for overnight at 4°C. Primary antibody \nconcentrations were as follows: anti -KAT7 (rabbit, 1:1000, Cell Signaling Technology, \n#58418), anti-JADE2 (rabbit, 1:2000, Proteintech, 11513 -1-AP), anti-iNOS (rabbit, 1:3000, \nGeneTex, GTX130246), anti -Iba1 (rabbit, 1:1000, Wako, #019 -19741), anti -p65 (rabbit, \n1:1000, Cell Signaling Technology, #8242), anti -phospho-p65 (Ser536) (rabbit, 1:1000, \nInvitrogen, #MA5-15160), anti-CMPK2 (rabbit, 1:1000, Novus Biologicals, # \tNBP1-80653), \nanti-GAPDH (mouse, 1:5000, Proteintech, 60004-1-lg), anti-β-actin (mouse, 1:5000, \nProteintech, 66009-1-lg), anti-H3K14ac (rabbit, 1:2000, Millipore, #07-353), anti-H3 (rabbit, \n1:3000, Proteintech, 17168 -1-AP). After wash, the membranes were incubated horseradish \nperoxidase (HRP) -conjugated secondary antibody (Cytiva, 1:5000). Immunoreactive bands \nwere visualized using Western Chemiluminescent HRP Substrate (Millipore, #WBKLS0500) \nand analyzed with ImageJ. \n \nReal-time qPCR \nTotal RNA was isolated from samples with TRIzol  reagents (Invitrogen) and was reverse \ntranscribed into cDNA using the HiScript III RT SuperMix for qPCR (+gDNA wiper) kits \n(R323-01, Vazyme). Relative quantitation was determined using the QuantStudio 6 Flex \ndetection system (Applied Biosystems) that measures real-time SYBR green fluorescence and \nthen calculated by means of the comparative Ct method (2−ΔΔCt) with the expression of Gapdh \nor β-actin as an internal control. The sequences of primers used are provided in Table S3. \n \nEdU staining \nTo measure newly synthesized mtDNA in primary microglia, in the presence of 10 µM EdU, \nthe cells were treated with LPS (0.2 µg/mL) or oligomeric Aβ42 (1 µM) for 6 hours and then \nwere incubated with MitoTracker (250 nM, Invitrogen) for 30 minutes. After cell fixation, \npermeabilization and blocking, EdU staining was performed according to the manufacturer’s \nprotocol using a Click -iT EdU Alexa Fluor 488 Imaging Kit (Invitrogen). The nucleus was \nstained with DAPI for 5 minutes. Images were collected with a Zeiss LSM 900 confoc al \nmicroscope and analyzed using ImageJ software (NIH). \n \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 11\t\nImmunofluorescence \nFor cells immunofluorescence, primary microglia were washed with PBS and fixed with 4% \nPFA for 20 minutes at room temperature. After PBS wash, cells were permeabilized with 0.2% \nTriton X-100 and blocked with blocking buffer (2% donkey serum plus 1% BSA in PBS). Cells \nwere incubated with primary antibodies overnight at 4°C. On the next day, cells were washed \nwith PBS and incubated with secondary antibodies (1:100, Jackson ImmunoResea rch) for 1 \nhour at room temperature.  \n \nFor immunofluorescence of mouse brain cryosections, anesthetized mice were perfused \ntranscardially with PBS, followed by 4% cold PFA in PBS. Brains were removed and fixed in \n4% PFA at 4°C overnight. After dehydration by 30% sucrose, brains were embedded in OCT \n(Tissue-Tek) and cut into 30-µm-thick sections on cryostat microtome (Leica). Sections were \npermeabilized and blocked with 0.3% Triton X -100 and 5% donkey serum in PBS for 1 hour \nat room temperature, and incubate d with primary antibodies at 4°C overnight. After washing \nthree times with PBS, slices were incubated with secondary antibodies (1:100, Jackson \nImmunoResearch) for 2 hours at room temperature.  \n \nFor immunofluorescence of formalin-fixed paraffin-embedded (FFPE) human patient tissues, \nthe brain tissue sections were deparaffinized in a 60°C oven for 2 hours, followed by xylene \nwashes twice, each for 10 minutes at room temperature. Tissues were then rehydrated in a \ngraded series of ethanol washes. Slides were rinsed with deionized water twice and transferred \ninto sodium citrate buffer (10 mM, adjust pH to 6.0) for antigen retrieval at 120  °C for 20 \nminutes. After cooling down to room temperature, sections were washed with PBS and \npermeabilized with 0.3% Triton X -100 in PBS for 20 minutes at room temperature, then \nblocked with blocking buffer (20% donkey serum plus 1%BSA in PBS) for 1 hour at room \ntemperature, and incubated with primary antibodies at 4°C overnight. On the next day, samples \nwere incubated with seco ndary antibodies (1:100, Jackson ImmunoResearch) for 2 hours at \nroom temperature. After washing three times with PBS, tissues were incubated with 0.1% \nSudan Black B in 70% ethanol for 30 minutes to quench autofluorescence. \n \nAfter samples were stained with DAPI and washed with PBS, samples were mounted using an \naqueous mounting medium (Aqua -Poly/Mount, Polysciences). Images were obtained with \nZeiss LSM900 confocal microscope  and analyzed with ImageJ. Primary antibody \nconcentrations were as follows: anti-Iba1 (goat, 1:200, Novus Biologicals, NB100-1028), anti-\nH3K14ac (rabbit, 1:100, Millipore, #07 -353), anti-Aβ (mouse, 1:200, Biolegend, #803004), \nanti-phospho-TBK1 (rabbit, 1:100, Cell Signaling Technology, #5483), anti -phospho-IRF3 \n(rabbit, 1:100, Cell Signaling Technology, #4947), anti-GFAP (mouse, 1:400, Invitrogen, 14-\n9892-82), anti-NeuN (mouse, 1:200, Millipore, MAB377). \n \nPreparation of oligomeric Aβ1-42 \nHFIP (hexofluoro-isopropanol) treated human Aβ1-42 peptides (rPeptide, A-1163) were first \ndissolved in dimethyl sulfoxide (DMSO) to a final  concentration of 5 mM, and this solution \nwas then diluted with cold phenol-free basal culture media to a final concentration of 250 µM.  \nOligomeric Aβ1-42 was prepared by incubation for 24 h at 4 °C, and this solution was aliquoted \nand stored at -80 °C before use.  \n \nThioflavin S (TS) staining \nAβ plaques were labeled by Thioflavin S staining on brain sections that were stained with 0.01% \nthioflavin S (T1892, Sigma) in 50% ethanol for 10 min. Then, sections were washed twice with \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 12\t\n50% ethanol and three times with PBS. Brain sections were mounted for imaging and analyzed \nusing a Zeiss LSM 900 confocal microscope. \n \nMeasurement of cytosolic mtDNA \nMicroglia were resuspended in digitonin buffer (50 mM HEPES, pH 7.4, 150 mM NaCl, and \n25 µg/ml digitonin) and incubated for 10 min at room temperature, followed by centrifugation \nat 2000g for 10 min at 4 °C. The supernatant containing cytosolic mtDNA (cmtDNA) was used \nfor qPCR. The pellet was used for nuclear DNA extraction with QIAamp DNA Mini Kit \n(Qiagen) according to the manufacturer’s instructions. The cmtDNA in the supernatant was \nnormalized to the nuclear DNA (B2m gene) in the pellet for each sample. D -loop and Nd4 \nwere used to assess mtDNA expression. B2m and Tert was used to assess nuclear DNA \nexpression. The sequences of primers used are provided in Table S4. \n \nRNA sequencing \nThree biological replicates were sequenced per group. For each sample, RNA was ext racted \nfrom BV2 cells with TRIzol reagents (Invitrogen). High-throughput RNA sequencing (RNA-\nseq) was performed by Illumina NovaSeq 6000 at Novogene (CA, USA). The raw sequencing \ndata were aligned to the mouse preference genome (GRCm39, mm39) using HISAT2 (v2.0.5). \nReads on each GENCODE annotated gene were counted using HTSeq, and then differential \ngene expression analysis was performed using DESeq2 R package. GO pathway analysis was \nconducted with DAVID tools (https://davidbioinformatics.nih.gov/).  \n \nCleavage Under Targets & Tagmentation (CUT&Tag) \nCUT&Tag was performed with Hyperactive Universal CUT&Tag Assay Kit for for Illumina \nPro (TD904, Vazyme) according to the manufacturer’s instructions. In brief, BV2 cells were \ncollected and counted same number for each group. Nuclei were isolated from BV2 cells and \nbounded to Concanavalin A (ConA)-coated beads. Subsequently, Nuclei were resuspended in \nantibody buffer and incubated with primary antibodies against H3K14ac (rabbit, Cell Signaling \nTechnology, #7627) and IgG (rabbit, Cell Signaling Technology, #66362) at 4  °C overnight. \nOn the next day, samples were incubated with goat anti-rabbit secondary antibodies (1:50, Cell \nSignaling Technology, #35401). The samples were incubated with pA/G -Tn5 transposase. \nAfter transposon activation and fragmentation, 0.5 pg Spike-in DNA was added to each sample \nand total DNA was isolated, amplified, and purified to constr uct library. The library for \nsequencing was constructed with TruePrep Index Kit V2 for Illumina (TD202, Vazyme) and \nVAHTS DNA Clean Beads (N411, Vazyme) were used for purification steps. The library was \nsequenced on an Illumina NovaSeq (PE 150) at Novogene . Raw sequencing reads were \ntrimmed using Cutadapt 5.0 (https://cutadapt.readthedocs.io/en/stable/). Trimmed reads were \nthen aligned to the mouse reference genome mm10 with the Spike-in sequence using Bowtie2 \n(version 2.3.5.1). Bam files with low-quality reads were filtered and duplicates were removed \nusing Samtools v1.18. Reads were then normalized to Spike-in using Bedtools v2.31.0. Peaks \nwere then called with SEACR v1.3. Differential peak analysis was analyzed by MAnorm2 and \nannotated by CHIPseeker with a p<0.05 cutoff. \n \nChIP-qPCR \nChIP experiments were performed according to the procedure described previously 53. BV2 \ncells were fixed with 1% formaldehyde for 15 min at room  temperature. The fixed cells were \nlysed in lysis buffer (1% SDS, 5 mM EDTA, 50 mM Tris -HCl, pH 8.1) containing  protease \ninhibitor cocktail. The lysates were then sonicated to generate chromatin fragments of ~500 bp \nin length. Cell debris was removed by centrifugation and supernatant were collected. A dilution \nbuffer (150 mM NaCl, 2 mM EDTA, 1% Triton X -100, and 20 mM Tris -HCl, pH 8.1) \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 13\t\ncontaining protease inhibitor cocktail was subsequently applied (1:9 ratio) and the chromatin \nsolution (40 µl aliquot as the input) was then incubated with specific antibodies (2 µg) at 4°C \novernight with mild rotation. 30 µl Protein A magnetic beads (Invitrogen) were added for \nincubation of 2 hours. Beads were sequentially washed with the following buffers: TSE I (150 \nmM NaCl, 2 mM EDTA, 0.1% SDS, 1% Triton X-100, 20 mM Tris-HCl, pH 8.1), TSE II (500 \nmM NaCl, 2 mM EDTA, 0.1% SDS, 1% Triton X -100, 20 mM Tris-HCl, pH 8.1), buffer III \n(0.25 M LiCl, 1% Nonidet P -40, 1 mM EDTA, 1% sodium deoxycholate, and 10 mM Tris -\nHCl, pH 8.1), a nd Tris-EDTA buffer. The input and the precipitated DNA -protein complex \nwere de-crosslinked at 65°C for 12 hours in elution buffer (1% SDS, 0.1 M NaHCO 3) with \nRNase A and Proteinase K. Then DNA was purified using QIAquick PCR Purification Kit \n(Qiagen). Quantification of the precipitated DNA fragments were performed with real -time \nPCR using primers listed in Table S5. \n \nIntracerebroventricular (ICV) injection \n5-month-old 5×FAD mice were used for ICV injection of WM-3835 (S9805, Selleck). Briefly, \n20 mice were randomly separated into two groups (WM-3835 and vehicle, 10 mice per group), \nthen deeply anesthetized  with isoflurane and immobiliz ed using a stereotactic device.  To \nimplant osmotic pumps in the mice, osmotic pumps (1004W  for 4 weeks infusion, RWD) \nmatched with Brain infusion kit (Bic -3, RWD) were loaded according to the manufacturer’s \ninstructions. 100 µL of vehicle (5% DMSO, 40% PEG300, and 55% saline ) or WM-3835 (1 \nmM) was filled in the osmotic pump, and the Bic-3 kit/tubing (2 cm) was backfilled before the \ntwo parts were connected. In addition to the incision on the scalp, the pocket for the osmotic \npump was obtained by stretching the space between the skin and the muscle in the back with \nsterile forceps. The detachable top part of the infusion cannula was held with a holder. A 0.5-\n1 mm burr hole was drilled in the skull, and the cannula tip was gently implanted into the lateral \nventricle ( coordinates, bregma: anterior/posterior,  -0.5 mm; medial/lateral, 1.0 mm; and \ndorsal/ventral: -2.3 mm). The os motic pump was slowly positioned in the pocket under the \nback skin simultaneously. The position of the cannula was secured with instant adhesives, and \nthe skin was sutured with suture thread. ICV infusion was performed for four weeks. Then, the \nmice were sacrificed and processed for pathology analyses. \n \nRNAscope in situ hybridization \nFixed brains were embedded in OCT (Tissue -Tek) and sectioned at a thickness of 14 µm. \nRNAscope Multiplex Fluorescent Reagent Kit v2 (ACD, Advanced Cell Diagnostics) was used \nfollowing the manufacturer’s manual. Probe targeting mouse  Kat7 (#1126701) or  Jade2 \n(#1725601) was purchased from ACD. Images were collected with a Zeiss LSM 900 confocal \nmicroscope and analyzed using ImageJ software (NIH). \n \nAcute brain slice electrophysiology \n6 months old mice were anesthetized with isoflurane, and then perfused intracardially with ice-\ncold oxygenated cutting solution containing (in mM ): 110 choline chloride, 2.5 KCl, 1.25 \nNaH2PO4, 25 NaHCO3, 0.5 CaCl2, 7 MgCl2, 10 glucose, saturated with 95% O2 / 5% CO2. The \nbrain was removed rapidly and immersed in ice -cold oxygenated cutting solution. Transverse \nhippocampal slices (35 0 µm) were cut i n the cutting solution using a vibratome (VT -1200S, \nLeica) and transferred to artificial cerebrospinal fluid (aCSF) containing (in mM): 125 NaCl, \n2.5 KCl, 1.25 NaH2PO4, 25 NaHCO3, 2 CaCl2, 2 MgCl2, 10 glucose, saturated with 95% O2 / \n5% CO 2. The slices wer e recovered for 20 min at 35 °C and then maintained at room \ntemperature for 1 hour. Slices were subsequently transferred to a submerged recording chamber \ncontaining aCSF solution maintained at 34 °C. Picrotoxin (100 µM) was added to block \ninhibitory transm ission. mEPSCs were recorded at a holding potential of -70 mV in the \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 14\t\npresence of 1 µM tetrodotoxin (TTX). fEPSPs were evoked in the CA1 stratum radiat um by \nstimulating the Schaffer collateral with a concentric bipolar electrode and recorded with a glass \npipette (1-3 MΩ) filled with aCSF.  The stimulus intensity was adjusted to evoke 40%-50% of \nthe maximal response. LTP was induced by theta burst stimulation (TBS) consisting of two \ntrains of 5 bursts at 5 Hz, and each burst contained 4 pulses at 100 Hz. Recordings were made \nwith MultiClamp 700B amplifier (Molecular Devices) and data acquisition was performed with \npClamp 10.7 software (Molecular Devices). \n \nMorris water maze test  \nMorris water maze tests were performed at Behavioral Core of Johns Hopkins University. In \nbrief, we used a maze consisted of a round pool (diameter, 120 cm) filled with water that was \nat 24 °C and made opaque with nontoxic white paint. A circular plastic platform (diameter, 10 \ncm) was placed at the center of the target quadrant and submerged 1 cm below the surface of \nthe water. Four local cues were provided to allow spatial map generation. In brief, we trained \nthe mice for four trials per day with different start points for five consecutive days. Mice were \ngently placed into the water facing the wall of the pool and allowed to freely explore the whole \nmaze for 1 min. Mice were then guided to the rescue platform if they did not find it. Mice were \nallowed to take a rest on the platform for 10 s and then retrained from a different start position \nwith the same procedure. The latency for each animal to find the platform (at least 3 s stay) \nwas recorded. On day 6, the platform was removed, and animals searched freely for 1 min \nstarting from the opposite quadrant. The entries into the platform area, total time spent in the \ntarget quadrant, and the total distance travels were recorded using the ANY-maze software. \n \nStatistical analysis \nStatistical analyses were performed using GraphPad Prism 9 (GraphPad Software, CA). Before \nstatistical analysis, variation within each group of data and the assumptions of the tests were \nchecked. For in vitro experiments, the cells were evenly suspended and randomly distributed \nin each well tested. For in vivo experiments, the animals were distributed into various treatment \ngroups randomly. Comparisons between two groups were made using unpaired Student’s two-\ntailed t test or Mann-Whitney test. Comparisons among three or more groups were made using \none- or two-way ANOVA followed by Bonferroni’s post hoc test. The significance level was \nset at p < 0.05. Test statistics, n numbers, and p values are indicated in the figure legends. All \ndata are presented as mean ± SEM. \n \nAcknowledgments: We thank Dr. Tony Wyss-Coray for sharing the BV2 cell line, Dr. Rong \nWu for kindly providing technical training in immunofluorescence of  human patient tissues , \nDr. Chaohua Jiang for the technical assistance on recording, and the members of Dr. Qiu lab \nfor valuable discussions.  \n \nFunding: This work was supported by NIH grants R35GM124824, R01NS118014, and \nRF1NS134549 (Z.Q.) and RF1NS113820, RF1NS127925, and R01AG078948 (S.S.). Z.Q. \nwas also supported by the KAT6 Foundation, the American Heart Association Established \nInvestigator A ward, McKnight Scholar Award, Klingenstein -Simon Scholar Award, Sloan \nResearch Fellowship in Neuroscience, and Randall Reed Scholar Award.. Y.Y. was supported \nby the Toffler Scholar Award. \n \nAuthor contributions : Y.L. initiated the project and performed the majority of the  \nexperiments. M.F. performed LTP recordings and osmotic pump implantation. Y.Y. and H.Y.C. \nanalyzed the sequencing data. H.Y.C. performed the flow cytometry of isolated microglia. S.S. \nprovided the human brain samples. Y.L., M.F., Y.Y., H.Y.C., and Z.Q. analyzed and \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 15\t\ninterpreted the results. Y.L. and Z.Q. designed the study and wrote the paper with input from \nall authors. \n \nCompeting interests: The authors declare that they have no competing interests. \n \n \nREFERENCES \n \n1.\t Chen,\tS.,\tCao,\tZ.,\tNandi,\tA.,\tCounts,\tN.,\tJiao,\tL.,\tPrettner,\tK.,\tKuhn,\tM.,\tSeligman,\tB., \t\nTortorice,\tD.,\tVigo,\tD.,\tet\tal.\t(2024).\tThe\tglobal\tmacroeconomic\tburden\tof\tAlzheimer's\t\ndisease\t and\t other\t dementias:\t estimates\t and\t projections\t for\t 152\t countries\t or\t\nterritories.\tLancet\tGlob\tHealth\t12,\te1534-e1543.\t10.1016/S2214-109X(24)00264-X.\t\n2.\t Knopman,\tD.S.,\tAmieva,\tH.,\tPetersen,\tR.C.,\tChetelat,\tG.,\tHoltzman,\tD.M.,\tHyman,\tB.T.,\t\nNixon,\t R.A.,\t and\t Jones,\t D.T.\t (2021).\t Alzheimer\t disease.\t Nat\t Rev\t Dis\t Primers\t7,\t 33.\t\n10.1038/s41572-021-00269-y.\t\n3.\t Self,\t W.K.,\t and\t Holtzman,\t D.M.\t (2023).\t Emerging\t diagnostics\t and\t therapeutics\t for\t\nAlzheimer\tdisease.\tNat\tMed\t29,\t2187-2199.\t10.1038/s41591-023-02505-2.\t\n4.\t Shi,\t F.D.,\t and\t Yong,\t V.W.\t (2025).\t Neuroinflammation\t across\t neurological\t diseases.\t\nScience\t388,\teadx0043.\t10.1126/science.adx0043.\t\n5.\t Heneka,\tM.T.,\tvan\tder\tFlier,\tW.M., \tJessen,\tF.,\tHoozemanns,\tJ.,\tThal,\tD.R.,\tBoche,\tD.,\t\nBrosseron,\t F.,\t Teunissen,\t C.,\t Zetterberg,\t H.,\t Jacobs,\t A.H.,\t et\t al.\t (2025).\t\nNeuroinflammation\t in\t Alzheimer\t disease.\t Nat\t Rev\t Immunol\t 25,\t 321 -352.\t\n10.1038/s41577-024-01104-7.\t\n6.\t van\tder\tFlier,\tW.M.,\tand\tHene ka,\tM.T.\t(2025).\tZooming\tin\ton\tbrain\tinflammation\tin\t\nAlzheimer's\tdisease.\tBrain\t148,\t1-2.\t10.1093/brain/awae394.\t\n7.\t Karch,\tC.M.,\tand\tGoate,\tA.M.\t(2015).\tAlzheimer's\tdisease\trisk\tgenes\tand\tmechanisms\t\nof\tdisease\tpathogenesis.\tBiol\tPsychiatry\t77,\t43-51.\t10.1016/j.biopsych.2014.05.006.\t\n8.\t Keren-Shaul,\t H.,\t Spinrad,\t A.,\t Weiner,\t A.,\t Matcovitch-Natan,\t O.,\t Dvir-Szternfeld,\t R.,\t\nUlland,\t T.K.,\t David,\t E.,\t Baruch,\t K.,\t Lara-Astaiso,\t D.,\t Toth,\t B.,\t et\t al.\t (2017).\t A\t Unique\t\nMicroglia\tType\tAssociated\twith\tRestricting\tDevelopment\tof\tAlzheimer's\tDisease.\tCell\t\n169,\t1276-1290\te1217.\t10.1016/j.cell.2017.05.018.\t\n9.\t Eskandari-Sedighi,\t G.,\t and\t Blurton-Jones,\t M.\t (2023).\t Microglial\t APOE4:\t more\t is\t less\t\nand\tless\tis\tmore.\tMol\tNeurodegener\t18,\t99.\t10.1186/s13024-023-00693-6.\t\n10.\t Leng,\t F.,\t and\t Edison,\t P.\t (2021).\t Neuroinflammation\t and\t microglial\t activation\t in\t\nAlzheimer\t disease:\t where\t do\t we\t go\t from\t here?\t Nat\t Rev\t Neurol\t 17,\t 157 -172.\t\n10.1038/s41582-020-00435-y.\t\n11.\t Zhong,\tF.,\tLiang,\tS.,\tand\tZhong,\tZ.\t(2019).\tEmerging\tRole\tof\tMitochondrial\tDNA\tas\ta\t\nMajor\t Driver\t of\t Inflammation\t and\t Disease\t Progression.\t Trends\t Immunol\t40,\t 1120-\n1133.\t10.1016/j.it.2019.10.008.\t\n12.\t Riley,\tJ.S.,\tand\tTait,\tS.W.\t(2020).\tMitocho ndrial\tDNA\tin\tinflammation\tand\timmunity.\t\nEMBO\tRep\t21,\te49799.\t10.15252/embr.201949799.\t\n13.\t Marchi,\t S.,\t Guilbaud,\t E.,\t Tait,\t S.W.G.,\t Yamazaki,\t T.,\t and\t Galluzzi,\t L.\t (2023).\t\nMitochondrial\t control\t of\t inflammation.\t Nat\t Rev\t Immunol\t 23,\t 159 -173.\t\n10.1038/s41577-022-00760-x.\t\n14.\t Gulen,\tM.F.,\tSamson,\tN.,\tKeller,\tA.,\tSchwabenland,\tM.,\tLiu,\tC.,\tGluck,\tS.,\tThacker,\tV.V.,\t\nFavre,\t L.,\t Mangeat,\t B.,\t Kroese,\t L.J.,\t et\t al.\t (2023).\t cGAS-STING\t drives\t ageing-related\t\ninflammation\t and\t neurodegeneration.\t Nature\t 620,\t 374-380.\t 10.1038/s41586-023-\n06373-1.\t\n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 16\t\n15.\t Xie,\tX.,\tMa,\tG.,\tLi,\tX.,\tZhao,\tJ.,\tZhao,\tZ.,\tand\tZeng,\tJ.\t(2023).\tActivation\tof\tinnate\timmune\t\ncGAS-STING\t pathway\t contributes\t to\t Alzheimer's\t pathogenesis\t in\t 5xFAD\t mice.\t Nat\t\nAging\t3,\t202-212.\t10.1038/s43587-022-00337-2.\t\n16.\t Udeochu,\t J.C.,\t Amin,\t S.,\t Huang,\t Y.,\t Fan,\t L.,\t Torres,\t E.R.S.,\t Carling,\t G.K.,\t Liu,\t B.,\t\nMcGurran,\t H.,\t Coronas -Samano,\t G.,\t Kauwe,\t G.,\t et\t al.\t (2023).\t Tau\t activation\t of\t\nmicroglial\tcGAS-IFN\treduces\tMEF2C -mediated\tcognitive\tresilience.\tNat\tNeurosci\t 26,\t\n737-750.\t10.1038/s41593-023-01315-6.\t\n17.\t Decout,\t A.,\t Katz,\t J.D.,\t Venkatraman,\t S.,\t and\t Ablasser,\t A.\t (2021).\t The\t cGAS-STING\t\npathway\tas\ta\ttherapeutic\ttarget\tin\tinflammatory\tdiseases.\tNat\tRev\tImmunol\t 21,\t548-\n569.\t10.1038/s41577-021-00524-z.\t\n18.\t Chung,\tS.,\tJeong,\tJ.H.,\tPark,\tJ.C.,\tHan,\tJ.W.,\tLee,\tY.,\tKim,\tJ.I.,\tand\tMook-Jung,\tI.\t(2024).\t\nBlockade\tof\tSTING\tactivation\talleviates\tmicroglial\tdysfunction\tand\ta\tbroad\tspectrum\t\nof\t Alzheimer's\t disease\t pathologies.\t Exp\t Mol\t Med\t56,\t 1936-1951.\t 10.1038/s12276-\n024-01295-y.\t\n19.\t Borrelli,\t E.,\t Nestler ,\t E.J.,\t Allis,\t C.D.,\t and\t Sassone -Corsi,\t P.\t (2008).\t Decoding\t the\t\nepigenetic\t language\t of\t neuronal\t plasticity.\t Neuron\t 60,\t 961 -974.\t\n10.1016/j.neuron.2008.10.012.\t\n20.\t Sweatt,\t J.D.\t (2013).\t The\t emerging\t field\t of\t neuroepigenetics.\t Neuron\t80,\t 624-632.\t\n10.1016/j.neuron.2013.10.023.\t\n21.\t Cavalli,\t G.,\t and\t Heard,\t E.\t (2019).\t Advances\t in\t epigenetics\t link\t genetics\t to\t the\t\nenvironment\tand\tdisease.\tNature\t571,\t489-499.\t10.1038/s41586-019-1411-0.\t\n22.\t Graff,\t J.,\t and\t Tsai,\t L.H.\t (2013).\t Histone\t acetylation:\t molecular\t mnemonics\t on\tthe\t\nchromatin.\tNat\tRev\tNeurosci\t14,\t97-111.\t10.1038/nrn3427.\t\n23.\t Yokoyama,\tA.,\tNiida,\tH.,\tKutateladze,\tT.G.,\tand\tCote,\tJ.\t(2024).\tHBO1,\ta\tMYSTerious\t\nKAT\t and\t its\t links\t to\t cancer.\t Biochim\t Biophys\t Acta\t Gene\t Regul\t Mech\t1867,\t 195045.\t\n10.1016/j.bbagrm.2024.195045.\t\n24.\t Pulido-Salgado,\tM.,\tVidal -Taboada,\tJ.M.,\tBarriga,\tG.G.,\tSola,\tC.,\tand\tSaura,\tJ.\t(2018).\t\nRNA-Seq\ttranscriptomic\tprofiling\tof\tprimary\tmurine\tmicroglia\ttreated\twith\tLPS\tor\tLPS\t\n+\tIFNgamma.\tSci\tRep\t8,\t16096.\t10.1038/s41598-018-34412-9.\t\n25.\t Oakley,\tH.,\tCo le,\tS.L.,\tLogan,\tS.,\tMaus,\tE.,\tShao,\tP.,\tCraft,\tJ.,\tGuillozet -Bongaarts,\tA.,\t\nOhno,\t M.,\t Disterhoft,\t J.,\t Van\t Eldik,\t L.,\t et\t al.\t (2006).\t Intraneuronal\t beta -amyloid\t\naggregates,\tneurodegeneration,\tand\tneuron\tloss\tin\ttransgenic\tmice\twith\tfive\tfamilial\t\nAlzheimer's\t disease\t mutations:\t potential\t factors\t in\t amyloid\t plaque\t formation.\t J\t\nNeurosci\t26,\t10129-10140.\t10.1523/JNEUROSCI.1202-06.2006.\t\n26.\t Srinivasan,\t K.,\t Friedman,\t B.A.,\t Etxeberria,\t A.,\t Huntley,\t M.A.,\t van\t der\t Brug,\t M.P.,\t\nForeman,\t O.,\t Paw,\t J.S.,\t Modrusan,\t Z.,\t Beach,\t T.G.,\t Serrano,\t G.E.,\t and\t Hansen,\t D.V.\t\n(2020).\t Alzheimer's\t Patient\t Microglia\t Exhibit\t Enhanced\t Aging\t and\t Unique\t\nTranscriptional\tActivation.\tCell\tRep\t31,\t107843.\t10.1016/j.celrep.2020.107843.\t\n27.\t MacPherson,\tL.,\tAnokye,\tJ.,\tYeung,\tM.M.,\tLam,\tE.Y.N.,\tChan,\t Y.C.,\tWeng,\tC.F.,\tYeh,\tP.,\t\nKnezevic,\t K.,\t Butler,\t M.S.,\t Hoegl,\t A.,\t et\t al.\t (2020).\t HBO1\t is\t required\t for\t the\t\nmaintenance\t of\t leukaemia\t stem\t cells.\t Nature\t577,\t 266-270.\t 10.1038/s41586-019-\n1835-6.\t\n28.\t Kueh,\tA.J.,\tDixon,\tM.P.,\tVoss,\tA.K.,\tand\tThomas,\tT.\t(2011).\tHBO1\tis\trequired\tfor\tH3K14\t\nacetylation\tand\tnormal\ttranscriptional\tactivity\tduring\tembryonic\tdevelopment.\tMol\t\nCell\tBiol\t31,\t845-860.\t10.1128/MCB.00159-10.\t\n29.\t Kueh,\tA.J.,\tBergamasco,\tM.I.,\tQuaglieri,\tA.,\tPhipson,\tB.,\tLi-Wai-Suen,\tC.S.N.,\tLonnstedt,\t\nI.M.,\t Hu,\t Y.,\t Feng,\t Z.P.,\t Woodruff,\t C.,\t May,\t R.E.,\t et\t al.\t (2023).\t Stem\t cell\t plasticity,\t\n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 17\t\nacetylation\tof\tH3K14,\tand\tde\tnovo\tgene\tactivation\trely\ton\tKAT7.\tCell\tRep\t42,\t111980.\t\n10.1016/j.celrep.2022.111980.\t\n30.\t Zhong,\tZ.,\tLiang,\tS.,\tSanchez-Lopez,\tE.,\tHe,\tF.,\tShalapour,\tS.,\tLin,\tX.J.,\tWong,\tJ.,\tDing,\tS.,\t\nSeki,\tE.,\tSchnabl,\tB.,\tet\tal.\t(2018).\tNew\tmitochondrial\tDNA\tsynthesis\tenables\tNLRP3\t\ninflammasome\tactivation.\tNature\t560,\t198-203.\t10.1038/s41586-018-0372-z.\t\n31.\t Xian,\tH.,\tand\tKarin,\tM.\t(2023).\tOxidized\tmitochondrial\tDNA :\ta\tprotective\tsignal\tgone\t\nawry.\tTrends\tImmunol\t44,\t188-200.\t10.1016/j.it.2023.01.006.\t\n32.\t Xian,\tH.,\tWatari,\tK.,\tSanchez -Lopez,\tE.,\tOffenberger,\tJ.,\tOnyuru,\tJ.,\tSampath,\tH.,\tYing,\t\nW.,\tHoffman,\tH.M.,\tShadel,\tG.S.,\tand\tKarin,\tM.\t(2022).\tOxidized\tDNA\tfragment s\texit\t\nmitochondria\t via\t mPTP -\t and\t VDAC -dependent\t channels\t to\t activate\t NLRP3\t\ninflammasome\t and\t interferon\t signaling.\t Immunity\t 55,\t 1370 -1385\t e1378.\t\n10.1016/j.immuni.2022.06.007.\t\n33.\t Miura,\tH.,\tQuadros,\tR.M.,\tGurumurthy,\tC.B.,\tand\tOhtsuka,\tM.\t(2018).\tEasi -CRISPR\tfor\t\ncreating\tknock-in\tand\tconditional\tknockout\tmouse\tmodels\tusing\tlong\tssDNA\tdonors.\t\nNat\tProtoc\t13,\t195-215.\t10.1038/nprot.2017.153.\t\n34.\t Sahasrabuddhe,\tV.,\tand\tGhosh,\tH.S.\t(2022).\tCx3Cr1 -Cre\tinduction\tleads\tto\tmicroglial\t\nactivation\tand\tIFN-1\tsignaling\tcaused\tby\tDNA\tdamage\tin\tearly\tpostnatal\tbrain.\tCell\tRep\t\n38,\t110252.\t10.1016/j.celrep.2021.110252.\t\n35.\t Yum,\tS.,\tLi,\tM.,\tFang,\tY.,\tand\tChen,\tZ.J.\t(2021).\tTBK1\trecruitment\tto\tSTING\tactivates\t\nboth\t IRF3\t and\t NF-kappaB\t that\t mediate\t immune\t defense\t against\t tumors\t and\t viral\t\ninfections.\tProc\tNatl\tAcad\tSci\tU\tS\tA\t118.\t10.1073/pnas.2100225118.\t\n36.\t Venegas,\tC.,\tKumar,\tS.,\tFranklin,\tB.S.,\tDierkes,\tT.,\tBrinkschulte,\tR.,\tTejera,\tD.,\tVieira -\nSaecker,\tA.,\tSchwartz,\tS.,\tSantarelli,\tF.,\tKummer,\tM.P.,\tet\tal.\t(2017).\tMicroglia-derived\t\nASC\t specks\t cross-seed\t amyloid-beta\t in\t Alzheimer's\t disease.\t Nature\t 552,\t 355-361.\t\n10.1038/nature25158.\t\n37.\t Hur,\tJ.Y.,\tFrost,\tG.R.,\tWu,\tX.,\tCrump,\tC.,\tPan,\tS.J.,\tWong,\tE.,\tBarros,\tM.,\tLi,\tT.,\tNie,\tP.,\t\nZhai,\t Y.,\t et\t al.\t (2020).\t The\t i nnate\t immunity\t protein\t IFITM3\t modulates\t gamma -\nsecretase\tin\tAlzheimer's\tdisease.\tNature\t586,\t735-740.\t10.1038/s41586-020-2681-2.\t\n38.\t Tsien,\tJ.Z.,\tChen,\tD.F.,\tGerber,\tD.,\tTom,\tC.,\tMercer,\tE.H.,\tAnderson,\tD.J.,\tMayford,\tM.,\t\nKandel,\t E.R.,\t and\t Tonegawa,\t S.\t (199 6).\t Subregion-\t and\t cell\t type -restricted\t gene\t\nknockout\tin\tmouse\tbrain.\tCell\t87,\t1317-1326.\t10.1016/s0092-8674(00)81826-7.\t\n39.\t Orihuela,\tR.,\tMcPherson,\tC.A.,\tand\tHarry,\tG.J.\t(2016).\tMicroglial\tM1/M2\tpolarization\t\nand\tmetabolic\tstates.\tBr\tJ\tPharmacol\t173,\t649-665.\t10.1111/bph.13139.\t\n40.\t Sangineto,\tM.,\tCiarnelli,\tM.,\tCassano,\tT.,\tRadesco,\tA.,\tMoola,\tA.,\tBukke,\tV.N.,\tRomano,\t\nA.,\t Villani,\t R.,\t Kanwal,\t H.,\t Capitanio,\t N.,\t et\t al.\t (2023).\t Metabolic\t reprogramming\t in\t\ninflammatory\t microglia\t indicates\t a\t potential\t way\t of\t ta rgeting\t inflammation\t in\t\nAlzheimer's\tdisease.\tRedox\tBiol\t66,\t102846.\t10.1016/j.redox.2023.102846.\t\n41.\t Natarajan,\tN.,\tFlorentin,\tJ.,\tJohny,\tE.,\tXiao,\tH.,\tO'Neil,\tS.P.,\tLei,\tL.,\tShen,\tJ.,\tOhayon,\tL.,\t\nJohnson,\t A.R.,\t Rao,\t K.,\t et\t al.\t (2024).\t Aberrant\t mitochondri al\t DNA\t synthesis\t in\t\nmacrophages\texacerbates\tinflammation\tand\tatherosclerosis.\tNat\tCommun\t 15,\t7337.\t\n10.1038/s41467-024-51780-1.\t\n42.\t Guan,\tX.,\tZhu,\tS.,\tSong,\tJ.,\tLiu,\tK.,\tLiu,\tM.,\tXie,\tL.,\tWang,\tY.,\tWu,\tJ.,\tXu,\tX.,\tand\tPang,\tT.\t\n(2024).\t Microglial\t CMPK2\t promo tes\t neuroinflammation\t and\t brain\t injury\t after\t\nischemic\tstroke.\tCell\tRep\tMed\t5,\t101522.\t10.1016/j.xcrm.2024.101522.\t\n43.\t Guan,\tJ.S.,\tHaggarty,\tS.J.,\tGiacometti,\tE.,\tDannenberg,\tJ.H.,\tJoseph,\tN.,\tGao,\tJ.,\tNieland,\t\nT.J.,\t Zhou,\t Y.,\t Wang,\t X.,\t Mazitschek,\t R.,\t et\t al.\t (2009).\t HDAC2\t negatively\t regulates\t\nmemory\tformation\tand\tsynaptic\tplasticity.\tNature\t459,\t55-60.\t10.1038/nature07925.\t\n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 18\t\n44.\t Mews,\t P.,\t Donahue,\t G.,\t Drake,\t A.M.,\t Luczak,\t V.,\t Abel,\t T.,\t and\t Berger,\t S.L.\t (2017).\t\nAcetyl-CoA\t synthetase\t regulates\t histone\t acetylati on\t and\t hippocampal\t memory.\t\nNature\t546,\t381-386.\t10.1038/nature22405.\t\n45.\t Liu,\tY.,\tFan,\tM.,\tYang,\tJ.,\tMihaljevic,\tL.,\tChen,\tK.H.,\tYe,\tY.,\tSun,\tS.,\tand\tQiu,\tZ.\t(2024).\t\nKAT6A\t deficiency\t impairs\t cognitive\t functions\t through\t suppressing\t RSPO2/Wnt\t\nsignaling\tin\thippocampal\tCA3.\tSci\tAdv\t10,\teadm9326.\t10.1126/sciadv.adm9326.\t\n46.\t Zhang,\t K.,\t Schrag,\t M.,\t Crofton,\t A.,\t Trivedi,\t R.,\t Vinters,\t H.,\t and\t Kirsch,\t W.\t (2012).\t\nTargeted\tproteomics\tfor\tquantification\tof\thistone\tacetylation\tin\tAlzheimer's\tdisease.\t\nProteomics\t12,\t1261-1268.\t10.1002/pmic.201200010.\t\n47.\t Plagg,\t B.,\t Ehrlich,\t D.,\t Kniewallner,\t K.M.,\t Marksteiner,\t J.,\t and\t Humpel,\t C.\t (2015).\t\nIncreased\tAcetylation\tof\tHistone\tH4\tat\tLysine\t12\t(H4K12)\tin\tMonocytes\tof\tTransgenic\t\nAlzheimer's\t Mice\t and\t in\t Human\t Patients.\t Curr\t Alzheimer\t Res\t 12,\t 752 -760.\t\n10.2174/1567205012666150710114256.\t\n48.\t Nativio,\t R.,\t Donahue,\t G.,\t Berson,\t A.,\t Lan,\t Y.,\t Amlie-Wolf,\t A.,\t Tuzer,\t F.,\t Toledo,\t J.B.,\t\nGosai,\t S.J.,\t Gregory,\t B.D.,\t Torres,\t C.,\t et\t al.\t (2018).\t Dysregulation\t of\t the\t epigenetic\t\nlandscape\t of\t normal\t aging \t in\t Alzheimer's\t disease.\t Nat\t Neurosci\t 21,\t 497 -505.\t\n10.1038/s41593-018-0101-9.\t\n49.\t Wang,\tW.,\tZheng,\tY.,\tSun,\tS.,\tLi,\tW.,\tSong,\tM.,\tJi,\tQ.,\tWu,\tZ.,\tLiu,\tZ.,\tFan,\tY.,\tLiu,\tF.,\tet\tal.\t\n(2021).\t A\t genome-wide\t CRISPR-based\t screen\t identifies\t KAT7\t as\t a\t driver\t of\t cellular\t\nsenescence.\tSci\tTransl\tMed\t13.\t10.1126/scitranslmed.abd2655.\t\n50.\t Xiao,\tY.,\tLi,\tW.,\tYang,\tH.,\tPan,\tL.,\tZhang,\tL.,\tLu,\tL.,\tChen,\tJ.,\tWei,\tW.,\tYe,\tJ.,\tLi,\tJ.,\tet\tal.\t\n(2021).\t HBO1\t is\t a\t versatile\t histone\t acyltransferase\t critical\t for\t promoter\t histone\t\nacylations.\tNucleic\tAcids\tRes\t49,\t8037-8059.\t10.1093/nar/gkab607.\t\n51.\t Zhou,\tY.,\tJia,\tK.,\tWang,\tS.,\tLi,\tZ.,\tLi,\tY.,\tLu,\tS.,\tYang,\tY.,\tZhang,\tL.,\tWang,\tM.,\tDong,\tY.,\tet\t\nal.\t (2023).\t Malignant\t progression\t of\t liver\t cancer\t progenitors\t requires\t lysine\t\nacetyltransferase\t 7 -acetylated\t and\t cytoplasm -translocated\t G\t protein\t GalphaS.\t\nHepatology\t77,\t1106-1121.\t10.1002/hep.32487.\t\n52.\t Marschallinger,\t J.,\t Iram,\t T.,\t Zardeneta,\t M.,\t Lee,\t S.E.,\t Lehallier,\t B.,\t Haney,\t M.S.,\t\nPluvinage,\t J.V.,\t Mathur,\t V.,\t Hahn,\t O.,\t Morgens,\t D.W.,\t et\t al.\t (2020).\t Lipid-droplet-\naccumulating\t microglia\t represent\t a\t dysfunctional\t and\t proinflammatory\t state\t in\t the\t\naging\tbrain.\tNat\tNeurosci\t23,\t194-208.\t10.1038/s41593-019-0566-1.\t\n53.\t Liu,\tY.,\tLai,\tS.,\tMa,\tW.,\tKe,\tW.,\tZhang,\tC.,\tLiu,\tS.,\tZhang,\tY.,\tPei,\tF.,\tLi,\tS.,\tY i,\tM.,\tet\tal.\t\n(2017).\tCDYL\tsuppresses\tepileptogenesis\tin\tmice\tthrough\trepression\tof\taxonal\tNav1.6\t\nsodium\tchannel\texpression.\tNat\tCommun\t8,\t355.\t10.1038/s41467-017-00368-z.\t\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 19\t\n \n \n \nFig. 1. Expression of the KAT7 complex is elevated in microglia from both 5×FAD mouse \nmodel and human AD patients . A, Heatmap of HATs components from public RNA -seq \ndataset (GSE90046, n=3 per group) in mouse primary microglia treated with LPS. B, Heatmap \nof HATs components from RNA-seq data (n=3 per group) in BV2 microglia treated with LPS. \nC, qPCR analysis of KAT7 complex in BV 2 cells treated with LPS. n=4 per group. Unpaired \nstudent’s t test. D, qPCR analysis of KAT7 complex in the isolated microglia from 6 -month-\nold 5×FAD mice. n=4 per group. Unpaired student’s t test. E, Left: Representative images of \nH3K14ac co-stained with Aβ plaques and microglia (Iba1) in the cortex region of 6-month-old \nWT and 5×FAD mice. Scale bar, 20 µm (left), 5 µm (right). Right: Quantification of H3K14ac \nintensity in microglia. n=100 cells from 4 mice per group . Mann -Whitney test. F, Left: \nRepresentative images of H3K14ac co -stained with Aβ plaques and microglia  in the frontal \nlobe region of AD patients and healthy controls. Scale bar, 6 µm. Right: Quantification of \nH3K14ac intensity in microglia. n=80 cells from 4 samples per group . Mann-Whitney test. \nWhite arrowheads indicate H3K14ac in microglia. *p<0.05, **p<0.01, ***p<0.001. Data are \npresented as mean ± SEM. \n \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 20\t\n \n \nFig. 2. KAT7 regulates LPS- and Aβ- induced inflammatory responses in microglia. A, \nGeneration of Kat7-KO BV2 cells. B, Western blot analysis of KAT7 and iNOS levels in Kat7-\nKO BV2 cells treated with LPS. C-D, qPCR and ELISA analysis of IL -6 levels in Kat7-KO \nBV2 cells treated with LPS. n=3. E, Western blot analysis of KAT7 overexpression in BV2 \ncells. F, ELISA analysis o f IL -6 secretion in BV2 cells treated with LPS. n=4. G, qPCR \nanalysis of Kat7 level in mouse primary microglia transfected with siRNA. n=4. H-I, qPCR \nanalysis of Nos2 and Il-6 levels in primary microglia treated with siRNA and LPS. n=3. J, \nELISA analysis of IL-6 secretion in primary microglia. n=3. K, qPCR analysis of Nos2 and Il-\n6 levels in primary microglia treated with WM -3835 and LPS. n=3. L, Schematic diagram \nshowing mouse primary microglia treated with WM -3835 and oligomeric Aβ42. M, qPCR \nanalysis of Il-6 levels in primary microglia treated with WM-3835 and oligomeric Aβ42. n=4. \nN-O, ELISA analysis of IL-6 secretion (6 h after Aβ42 treatment) and IL-1β (24 h after Aβ42 \ntreatment) in primary microglia treated with WM-3835 and Aβ42. n=4. **p<0.01, ***p<0.001. \nOne-way ANOVA test (G, K, and M-O). Two-way ANOVA test (C, D, F, and H-J). Data are \nmean±SEM. \n \n \n \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 21\t\n \nFig. 3. Integrated transcriptomic and epigenomic analysis identifies Cmpk2 as a key target \nof KAT7. A, Heatmap of RNA -seq in WT and Kat7-KO BV2 cells with or without LPS \nstimulation. n=3 replicates per group. B, Volcano plot showing differentially expressed genes \nbetween WT_LPS and WT BV2 cells. C, Volcano plot showing differentially expressed genes \nbetween Kat7-KO_LPS and WT_LPS BV2 cells. D, Venn diagram (left) of overlapping genes \namong downregulated in KO vs WT, downregulated in KO_LPS vs WT_LPS, and upregulated \nin WT_LPS vs WT. Right: GO pathway analysis of the 110 overlapped genes. E, Heatmap of \nCUT&Tag with IgG and H3K14ac antibodies  in WT and Kat7-KO BV2 cells treated with or \nwithout LPS stimulation. F, Top: Venn diagram showing overlapped genes of RNA -seq (110 \ngenes) and CUT&Tag (244 genes). Bottom: Heatmap of 17 overlapped genes in RNA-seq. G, \nRepresentative CUT&Tag tracks of H3K14 ac in Cmpk2. Green box indicated proximal \npromoter. TSS, transcriptional start site. H, qPCR analysis of Cmpk2 in BV2 cells treated with \nLPS. n=3. I, qChIP analysis of Cmpk2 promoter using the indicated antibodies in BV2 cells. \nn=3. J, qPCR analysis of Cmpk2 in primary microglia treated with Kat7 siRNA and oligomeric \nAβ42. n=3. **p<0.01, ***p<0.001. One-way ANOVA test. Data are mean±SEM. \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 22\t\n \n \nFig. 4.  Kat7 knockdown in microglia limits mtDNA replication and release through \nrepressing Cmpk2 transcription. A-B, Representative images ( A) and quantification ( B) of \nEdU-labelled newly synthesized mtDNA in primary microglia treated with LPS. Scale bar, 5 \nµm. n=20 cells from 3 replicates per group. C, Quantification of cytosolic mtDNA (cmtDNA) \nby qPCR (normalized to nuclear B2m DNA) in primary microglia treated with LPS plus ATP. \nD-loop indicated a specific fragment within mtDNA. n=3. D-E, Representative images (D) and \nquantification (E) of p -TBK1 in primary microglia treated with LPS plus ATP.  Scale bar, 5 \nµm. n=10 cells from 2 replicates per group. F-G, Representative images (F) and quantification \n(G) of p-IRF3 in nuclei of primary microglia treated with LPS plus ATP. Scale bar, 3 µm. n=10 \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 23\t\ncells from 2 replicates per group. H, ELISA analysis of IL-1β secretion from primary microglia \ntreated with LPS plus  ATP. n=3. **p<0.01, ***p<0.001. One -way ANOVA test ( E and G). \nTwo-way ANOVA test (B, C and H). Data are presented as mean ± SEM.  \n \n \n \n \n \n \n \n \n \n \n \n \n \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 24\t\n \nFig. 5. Microglia-specific Kat7 deletion inhibits neuroinflammation in 5×FAD mice. A, \nSchematic diagram showing the strategy for generating Kat7-specific deletion in microglia in \n5×FAD mice. B, Knockout efficiency of Kat7 in microglia was examined by western blot (left) \nand qPCR analysis (right). Unpaired student’s t test. C, qPCR analysis of Cmpk2 and Mx2 in \nmicroglia isolated from 6-month-old mice. n=4. Two-way ANOVA test. D, Quantification of \ncytosolic mtDNA by qPCR (normalized to nuclear B2m DNA) in microglia isolated from 6 -\nmonth-old mice. D-loop and Nd4 indicated a specific fragment within mtDNA. Tert and B2m \nindicated nuclear DNA. n=3. Two-way ANOVA test. E-F, Representative images of p-TBK1 \n(E) and p -IRF3 (F) staining in cortex region of 6-month-old AD mice. n=4 mice per group. \nScale bar, 10 µm. One-way ANOVA test. G, Quantification of p -TBK1 (E) and p -IRF3 (F) \nintensity in microglia of 6-month-old AD mice. n=4 mice per group. Unpaired student’s t test. \nH, ELISA analysis of IL-1β and IL -6 production in cortex region of 6-month-old mice . I, \nRepresentative images (left) and quantification (right) of Iba1+ microglia in 6-month-old mice. \nn=4 mice per group. Scale bar, 100 µm. Unpaired student’s t test. J, Immunoblot analysis of \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 25\t\nIba1 and p-p65 in the cortex of 6-month-old mice (left), and protein levels were normalized to \nβ-actin (right). n=4 mice per group. Two-way ANOVA test. *p<0.05, **p<0.01, ***p<0.001. \nData are presented as mean ± SEM.   \n \n \n \n \n \n \n \nFig. 6. Microglia-specific Kat7 deletion ameliorates Aβ pathology and improves cognitive \nfunction in 5×FAD mice. A, Representative images of TS staining in the brain sections of 6-\nmonth-old AD mice . Scale bar, 0.5 mm (left), 0.1  mm (right). TAM: tamoxifen. B, \nQuantification of TS-positive Aβ plaque number and area in 6-month-old AD mice. n=4 mice \nper group. One-way ANOVA test. C, TBS-induced LTP at Schaffer collateral to CA1 synapses \nin 6-month-old mice. Arrow indicates LTP induction . Sample traces represent fEPSP taken \nbefore (1) and 50 min after (2) TBS. D, Averaged fEPSP slopes during 50 to 60 min after the \nstimulation. n=10-11 slices from 5 mice per group. One -way ANOVA test. E, Time spent \nbefore reaching the hidden platform during training days in the Morris water maze test. Two -\nway ANOVA test. F-I, Representative traces (F), time spent in target quadrant (G), entries into \nplatform zone ( H) and swimming speed ( I) during the probe test. n=11 -15 mice per group. \nOne-way ANOVA test. *p<0.05, **p<0.01. ns, nonsignificant. Data are presented as mean ± \nSEM.   \n \n \n \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 26\t\n \n \nFig. 7. Pharmacological inhibition of KAT7 reduces neuroinflammation and Aβ burden \nin 5×FAD mice. A, qPCR analysis of Cmpk2 levels in primary microglia treated with WM-\n3835 and oligomeric Aβ42. n=4. B, Representative images (left) and quantification (right) of \nEdU-labelled newly synthesized mtDNA in primary microglia treated with WM-3835 and \nAβ42. Scale bar, 4 µm. n=30 cells from 3 replicates per group. C, Schematic diagram showing \nthe strategy for del ivering vehicle or WM -3835 into ICV via osmotic pump in 5 -month-old \n5×FAD mice. D-F, Representative images ( D) and quantification of H3K14ac intensity in \nmicroglia (E) and Iba1+ microglia number (F) in cortex of 6-month-old AD mice treated with \nvehicle or WM-3835. Scale bar, 20 µm. n=4 mice per group. G, Quantification of cytosolic \nmtDNA by qPCR (normalized to nuclear B2m DNA) in microglia isolated from 6 -month-old \nAD mice treated with vehicle or WM -3835. D-loop and Nd4 indicated a specific fragment \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint \n\n\t 27\t\nwithin mtDNA. B2m indicated nuclear DNA. H, Representative images (left) and \nquantification (right) of p-TBK1 in microglia of 6-month-old AD mice treated with vehicle or \nWM-3835. Scale bar, 10 µm. n= 4 mice  per group . I, Representative images (left) and \nquantification (right) of TS staining in the cortex and hippocampus of 6 -month-old AD mice \ntreated with vehicle or WM -3835. Scale bar, 0.4 mm (left), 0.1 mm ( middle and right). n=4 \nmice per group. J, Time spent before reaching the hidden platform during training days in the \nMorris water maze test. K-L, Time spent in different quadrants ( K) and swimming speed (L) \nduring the probe test. n=10 mice per group. *p<0.05, **p<0.01 , ***p<0.001. Unpaired \nstudent’s t test (E, F, H and L). One-way ANOVA test (A-B). Two-way ANOVA test (G, I, J \nand K). Data are mean±SEM. \n \n \n \n \n \n \n \n \n \n \nFig. 8 . Summary diagram . Our results  support a model in which KAT7 acts as a central \nepigenetic driver of neuroinflammation by promoting H3K14ac-dependent Cmpk2 expression. \nThrough this mechanism, KAT7 integrates nuclear and mitochondrial responses to amplify \nneuroinflammation, thereby contributing to Aβ accumulation and synaptic dysfunction in AD.    \n \n \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted February 20, 2026. ; https://doi.org/10.64898/2026.02.19.706884doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}