CoChat
Sign Up
Cited by 55Neuroinflammation and Neurodegeneration MechanismsAlzheimer's disease research and treatmentsDementia and Cognitive Impairment Research

PICALM Alzheimer’s risk allele causes aberrant lipid droplets in microglia

Open access
Kozlova, Alena, Zhang, Siwei, Sudwarts, Ari, Zhang, Hanwen, Smirnou, Stanislau, Byeon, Seul Kee, Thapa, Christina, Sun, Xiaotong +13 more·2025·Nature
DOI 10.1038/s41586-025-09486-xPMC PMC11142308

Published version of a preprint

Open this paper in CoChat — free account

Abstract

, the underlying disease mechanisms remain largely unclear. Determining causal disease variants and their LOAD-relevant cellular phenotypes has been a challenge. Here, using our approach for identifying functional GWAS risk variants showing allele-specific open chromatin, we systematically identified putative causal LOAD-risk variants in human induced pluripotent stem (iPS)-cell-derived neurons, astrocytes and microglia, and linked a PICALM LOAD-risk allele to a microglial-specific role of PICALM in lipid droplet (LD) accumulation. Allele-specific open-chromatin mapping revealed functional risk variants for 26 LOAD-risk loci, mostly specific to microglia. At the microglial-specific PICALM locus, the LOAD-risk allele of the single-nucleotide polymorphism rs10792832 reduced transcription factor (PU.1) binding and PICALM expression, impairing the uptake of amyloid beta (Aβ) and myelin debris. Notably, microglia carrying the PICALM risk allele showed transcriptional enrichment of pathways for cholesterol synthesis and LD formation. Genetic and pharmacological perturbations of microglia further established a causal link between reduced PICALM expression, LD accumulation and phagocytosis deficits. Our work elucidates the selective LOAD vulnerability in microglia at the PICALM locus through detrimental LD accumulation, providing a neurobiological basis that can be exploited for developing clinical interventions.

AI snapshot

Written from the paper’s full text.

TL;DR

Using allele-specific open chromatin mapping in human iPS-cell-derived brain cell types, this study identifies functional late-onset Alzheimer's disease (LOAD) risk variants for 26 loci, mostly microglia-specific, and links the PICALM risk allele of rs10792832 to reduced PU.1 binding, reduced PICALM expression, lipid droplet accumulation and impaired phagocytosis in microglia.

Key findings

  • ASoC mapping identified functional LOAD-risk variants for 26 loci, with 20 loci (38 ASoC/LOAD-risk SNPs) accounted for by iMGs; iMG open-chromatin regions were strongly enriched for LOAD heritability (enrichment fold = 13.6, P = 2.1 × 10⁻¹⁵) and iMG ASoC SNPs showed the strongest LOAD GWAS enrichment (35-fold, P = 1.04 × 10⁻³³).
  • The LOAD-risk allele G of rs10792832 disrupted a PU.1 binding motif, retaining about 40% of allele A's PU.1 binding, and reduced PICALM expression by about 50% in iMGs (but not iAst cells); lower PICALM expression was associated with the risk allele G in AD patients (P = 0.004).
  • Risk-allele iMGs showed 40–50% decreased Aβ–pHrodo phagocytosis and ~50% reduced myelin phagocytosis in PICALM-KO C20 cells; PICALM-CRISPRa rescued the phagocytic deficit.
  • Risk-allele iMGs showed a twofold to sevenfold increase in lipid droplets, a ~twofold increase in total triacylglycerols (32 of 55 altered lipid species were TGs, 5.6-fold enrichment, P < 2.6 × 10⁻¹⁶), >2-fold increase in ROS, threefold more peroxidated lipids, and 2.7–4.2-fold more lysosomes (4.3–5.7-fold for lysosome–LD colocalization).

Methods

  • ATAC–seq and allele-specific open chromatin (ASoC) mapping in human iPS-cell-derived microglia (iMG, n = 38), astrocytes (iAst, n = 18), and glutamatergic, GABAergic and dopaminergic neurons (62 iPS cell lines total), integrated with hMG caQTL/eQTL, ABC enhancer–promoter linkage and LOAD GWAS data.
  • CRISPR–Cas9 editing of rs10792832 (G/G→A/A) in two non-AD donor iPS lines (CD04, CD09), plus CRISPRa (PICALM overexpression), CRISPRoff (knockdown) and PICALM knockout in the C20 hMG line; assays included ChIP for PU.1, RNA-seq, lipidomics (LC–MS/MS), phagocytosis (Aβ- and myelin-pHrodo), LD/ROS/lysosome staining and flow cytometry.
  • Statistical analysis used linear mixed models, one-way ANOVA with Dunnett's or Tukey's correction, two-sided binomial tests for ASoC, and Benjamini–Hochberg FDR correction.

Limitations

  • The authors acknowledge a lack of strong correlation between the magnitude of PICALM alteration and its functional effects in CRISPR-engineered iMGs, possibly due to differentiation variability and nonlinear gene–phenotype relationships; they cannot formally exclude a long-range cis-effect of the risk variant on genes other than PICALM.
  • The authors note PICALM expression in AD patients may be influenced by cell type, MG state and disease stage, and that effects seen in iMGs may also occur in other myeloid cells/macrophages, warranting further investigation.
  • The authors note that although impaired phagocytosis implies a neuroprotective role in clearing Aβ or damaged myelin, MG phagocytosis in vivo may also be detrimental through excessive synaptic pruning.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s41586-025-09486-x

Cites

89 of 89 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

64 of 77 cited sources checked · 31 supported · 0 flagged · accuracy not yet measured

  • Study design and conductown data
    § Mapping functional LOAD-risk variants

    The iMG and iAst cells were epigenetically similar to hMGs and other reported iMGs and Ast cells.

    “Our iMG and iAst cells were epigenetically similar to hMGs6 and other reported iMGs22 and Ast (hAst)6 cells (Extended Data Fig. 1i).”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [6] Kosoy, R. et al. Genetics of the human microglia regulome refines Alzheimer’s disease risk loci. Nat. Genet. 54, 1145–1154 (2022). source

      The Results section, “Chromatin accessibility and gene expression in microglia” (page unavailable), confirms that this paper generated chromatin-accessibility data from primary human microglia, supporting its use as an hMG comparator. The cited paper does not establish that the citing authors’ iMG or iAst cells were epigenetically similar to those cells, and our copy does not identify an hAst epigenetic dataset from this paper. The comparison with other reported iMGs is attached to citation 22, not this source. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
    • cites [22] Novikova, G. et al. Integration of Alzheimer’s disease genetics and myeloid genomics identifies disease risk regulatory elements and genes. Nat. Commun. 12, 1610 (2021). source
  • Findingown data
    § 100 bp

    Lower PICALM expression was found in disease-associated MGs (DAMs) and LD-accumulating MGs (LDAMs) versus homeostatic MGs (P = 0.016).

    “Moreover, by recapitulating snRNA-seq analysis of AD brain MG states19, we found lower PICALM expression in disease-associated MGs (DAMs) and LD-accumulating MGs (LDAMs) (versus homeostatic MGs) (P = 0.016) (Extended Data Fig. 4c).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [19] Haney, M. S. et al. APOE4/4 is linked to damaging lipid droplets in Alzheimer’s disease microglia. Nature 628, 154–161 (2024). source
  • Premisecitation
    § 0 2 4 6

    ATP6AP2 encodes a protein essential for lysosomal degradative functions.

    “ATP6AP2, one of the most downregulated genes in risk-allele iMGs (Fig. 4b and Supplementary Table 13), encodes a protein that is essential for lysosomal degradative functions; Atp6ap2-mutant mice show increased triglycerides and LD in fat body cells48.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [48] Hirose, T. et al. ATP6AP2 variant impairs CNS development and neuronal survival to cause fulminant neurodegeneration. J. Clin. Invest. 129, 2145–2162 (2019). source
  • Findingown data
    § 0 2 4 6

    Three known LD suppressor genes (SLC33A1, MCOLN1 and GRN) all exhibit reduced expression in LOAD-risk-allele iMGs.

    “Notably, three known LD suppressor genes (SLC33A1, MCOLN1 and GRN)10 all exhibited reduced expression in LOAD-risk-allele iMGs (Fig. 5d), two of which are related to lysosomal function: MCOLN1 is a ROS sensor in lysosomes regulating autophagy49, and GRN encodes the lysosomal protein progranulin related to neurodegenerative diseases50.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [10] Marschallinger, J. et al. Lipid-droplet-accumulating microglia represent a dysfunctional and proinflammatory state in the aging brain. Nat. Neurosci. 23, 194–208 (2020). source

      In Results (CRISPR-Cas9 screen; page unavailable), the source identifies SLC33A1 and GRN as negative regulators of lipid-droplet formation: disrupting either increased lipid droplets in BV2 cells. The cited paper does not identify MCOLN1 as a lipid-droplet suppressor, so it supports only two of the three genes named. The reduced expression in LOAD-risk-allele iMGs is attributed to the citing paper’s Fig. 5d, not to this source. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
  • Findingown data
    § 0 2 4 6

    Blocking LD formation with TrC influences ROS levels in iMGs.

    “As seen in mouse microglia10, we further confirmed that LD accumulation influences ROS levels in iMGs by using TrC to block LD formation (Extended Data Fig. 7f–i).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [10] Marschallinger, J. et al. Lipid-droplet-accumulating microglia represent a dysfunctional and proinflammatory state in the aging brain. Nat. Neurosci. 23, 194–208 (2020). source
  • Classified as probably not a substantial claim. 57
  • May combine two claims. 26
  • The quote only partly states this claim. 3

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § 100 bp

    The functional linkage between rs6733839 and BIN1 has been well established.

    “including the two loci with the strongest associations with LOAD: the BIN1 locus for which the functional linkage between rs6733839 and BIN1 has been well established7,23”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [7] Yang, X. et al. Functional characterization of Alzheimer’s disease genetic variants in microglia. Nat. Genet. 55, 1735–1744 (2023). source

      The Results section ('AD variants affect multiple target genes in microglia'; page unavailable) reports that perturbing a candidate regulatory element at the BIN1 locus significantly affects nearby gene expression. That supports a functional link at the locus, but the copy does not identify rs6733839 as the responsible variant or establish its specific functional linkage to BIN1. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [23] Nott, A. et al. Brain cell type-specific enhancer–promoter interactome maps and disease-risk association. Science 366, 1134–1139 (2019). source

      The paper’s BIN1 enhancer discussion (indexed under Abstract; no page supplied) places rs6733839 in a microglia-specific enhancer linked to the BIN1 promoter and reports that deleting the enhancer sharply reduced BIN1 expression in microglia. That establishes a functional link between the enhancer and BIN1, but the paper does not test the effect of rs6733839 itself; it calls for further investigation of the risk allele’s function. Thus, saying the variant’s functional linkage is well established overstates this source. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Introduction

    APOE4 impairs neuron-glia lipid metabolism.

    “APOE4, the strongest genetic risk factor for LOAD, impairs neuron–glia lipid metabolism14,16”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [14] Qi, G. et al. ApoE4 impairs neuron-astrocyte coupling of fatty acid metabolism. Cell Rep. 34, 108572 (2021). source
    • cites [16] Liu, L., MacKenzie, K. R., Putluri, N., Maletić-Savatić, M. & Bellen, H. J. The glia-neuron lactate shuttle and elevated ROS promote lipid synthesis in neurons and lipid droplet accumulation in Glia via APOE/D. Cell Metab. 26, 719–737 (2017). source

  • May combine two claims.

    Findingcitation
    § 100 bp

    rs10792832 lies within a predicted binding site of PU.1, with the risk allele G predicted to disrupt PU.1's consensus binding motif.

    “The PICALM SNP rs10792832 lies within a predicted binding site of PU.1, a key myeloid-specific TF encoded by a strong LOAD-risk gene, SPI13,28, with the risk allele G predicted to disrupt PU.1's consensus binding motif (Fig. 2d).”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [3] Bellenguez, C. et al. New insights into the genetic etiology of Alzheimer’s disease and related dementias. Nat. Genet. 54, 412–436 (2022). source

      The GWAS meta-analysis table lists Alzheimer’s-associated PICALM and SPI1 loci, but the passages from our copy do not identify rs10792832 or address PU.1 binding at that variant. Searches for rs10792832, PU.1, and the predicted motif disruption found no passage establishing the claim. (LLM, conf=0.73)

      Experimental — accuracy not yet measured
    • cites [28] Andrews, S. J. et al. The complex genetic architecture of Alzheimer’s disease: novel insights and future directions. eBioMedicine https://doi.org/10.1016/j.ebiom.2023.104511 (2023). source

      The ‘Alzheimer's disease gene prioritization’ section (page unavailable) says that AD risk variants affect PU.1 binding at microglial regulatory elements. It does not identify rs10792832, place it in a predicted PU.1 binding site, or say that its G allele disrupts the consensus motif. (LLM, conf=0.82)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § 100 bp

    The directional effect of the AD risk allele of rs10792832 on PICALM expression was consistent with other hMG eQTL findings.

    “The directional effect of the AD risk allele of rs10792832 on PICALM expression was consistent with other hMG eQTL findings8,9,25.”
    Quote in the paperNot a substantial claim3 of 3 sources checked
    • cites [8] Lopes, K. P. et al. Genetic analysis of the human microglial transcriptome across brain regions, aging and disease pathologies. Nat. Genet. 54, 4–17 (2022). source

      The Results sections on genetic regulatory effects and disease loci identify PICALM as an AD-associated microglial eQTL and report that PICALM colocalization was shared with a previously published microglial dataset. They do not identify rs10792832 or state the direction of its AD risk allele’s effect on PICALM expression, so the specific directional-consistency claim is not established. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
    • cites [9] Bryois, J. et al. Cell-type-specific cis-eQTLs in eight human brain cell types identify novel risk genes for psychiatric and neurological disorders. Nat. Neurosci. 25, 1104–1112 (2022). source

      The 'Brain cell type eQTLs mediate brain disease associations' section reports a microglial PICALM co-localization signal and associates increased PICALM expression with decreased AD risk. The 'Mapping of risk variants to cell type regulatory elements' section identifies rs10792832 at a microglia-specific enhancer connected to PICALM, but our copy does not establish the expression effect of that particular risk allele or compare its direction with other human microglia eQTL findings. The indexed passages provide section labels but no specific page for these statements. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [25] Jang, B. et al. SingleBrain: a meta-analysis of single-nucleus eQTLs linking genetic risk to brain disorders. Preprint at medRxiv https://doi.org/10.1101/2025.03.06.25323424 (2025). source

      The Results section discusses AD-associated microglial eQTLs, but the passages found do not mention rs10792832 or PICALM. Searches for both terms and for a directional comparison with other human microglial eQTL findings found no passage that addresses the claim. (LLM, conf=0.72)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Introduction

    APOE4-associated glial LD accumulation modulates neuron function and brain memory.

    “and the APOE4-associated glial LD accumulation modulates neuron function and brain memory17–19.”
    Quote in the paperNot a substantial claim3 of 3 sources checked
    • cites [17] Blanchard, J. W. et al. APOE4 impairs myelination via cholesterol dysregulation in oligodendrocytes. Nature 611, 769–779 (2022). source

      The Introduction (page unavailable) reports increased lipid droplets in APOE4 oligodendroglia and reduced droplets, improved myelination, and improved learning and memory after treatment in APOE4 mice. The Discussion links cholesterol dysregulation to impaired myelination and cognition, but the paper does not establish that lipid-droplet accumulation itself modulates neuronal function or memory; the intervention also changes cholesterol transport. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [18] Victor, M. B. et al. Lipid accumulation induced by APOE4 impairs microglial surveillance of neuronal-network activity. Cell Stem Cell 29, 1197–1212 (2022). source

      The Results sections on microglial energetics and lipid-content modulation report increased lipid droplets in APOE4 microglia and show that their conditioned media reduced neuronal bursts; reducing the droplets preserved neuronal activity. The Discussion section on neuronal excitability only proposes a possible detriment to neural computations critical for learning and memory. It does not establish an effect on memory itself. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
    • cites [19] Haney, M. S. et al. APOE4/4 is linked to damaging lipid droplets in Alzheimer’s disease microglia. Nature 628, 154–161 (2024). source

      The ‘ACSL1 and triglycerides increase after fAβ challenge’ section reports greater lipid-droplet accumulation in APOE4/4 microglia. The ‘LD+ iMG induce pTau and apoptosis in neurons’ section reports that conditioned media from lipid-droplet-high microglia induced Tau phosphorylation and caspase activation in cultured neurons, supporting an effect on neurons. Our copy does not report an effect on brain memory; the neuronal effects were also observed with APOE3/3 microglia, so they are not shown to be unique to APOE4. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • Classified as probably not a substantial claim.

    Premisecitation
    § 100 bp

    PICALM has a role in clathrin-mediated endocytosis in non-MG cells.

    “and the role of PICALM in clathrin-mediated endocytosis in non-MG cells31,32”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [31] Ando, K. et al. PICALM and Alzheimer’s disease: an update and perspectives. Cells https://doi.org/10.3390/cells11243994 (2022). source
    • cites [32] Zhang, B. et al. Synaptic vesicle size and number are regulated by a clathrin adaptor protein required for endocytosis. Neuron 21, 1465–1475 (1998). source

      The Results passage on the Drosophila lap gene (page not provided) identifies LAP as an AP180-family protein with a human homolog, and the paper reports that loss of LAP impairs synaptic-vesicle endocytosis. It does not directly establish a role for human PICALM in clathrin-mediated endocytosis in non-MG cells; that part is an extrapolation from the homolog studied. (LLM, conf=0.82)

      Experimental — accuracy not yet measured

  • Classified as probably not a substantial claim.

    Premisecitation
    § Introduction

    Lipid droplets (LDs) have important roles in ageing and Alzheimer's disease (AD).

    “Moreover, LDs—cellular organelles containing lipids such as glycerolipids and cholesterol—were recently found to have important roles in ageing and Alzheimer's disease (AD)10–15.”
    Quote in the paperSubstantial claim5 of 6 sources checked
    • cites [10] Marschallinger, J. et al. Lipid-droplet-accumulating microglia represent a dysfunctional and proinflammatory state in the aging brain. Nat. Neurosci. 23, 194–208 (2020). source

      The Results and Discussion describe lipid-droplet-accumulating microglia in the aging brain, including impaired phagocytosis (Discussion, ‘Role of lipid droplets in the LDAM functional phenotype’; page unavailable). The Introduction notes historical observations of lipid-filled glial cells in dementia, but our copy does not establish an important role for lipid droplets in Alzheimer’s disease specifically. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [11] Lindner, K. et al. Isoform- and cell-state-specific lipidation of ApoE in astrocytes. Cell Rep. 38, 110435 (2022). source

      The Discussion links lipid-droplet accumulation in astrocytes to aging and describes a possible connection to late-onset AD. But the Limitations of the study says its proposed effects on brain aging and AD remain speculative, so this source does not establish that lipid droplets have important roles in both. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
    • cites [12] Sienski, G. et al. APOE4 disrupts intracellular lipid homeostasis in human iPSC-derived glia. Sci. Transl. Med. https://doi.org/10.1126/scitranslmed.aaz4564 (2021). source
    • cites [13] Moulton, M. J. et al. Neuronal ROS-induced glial lipid droplet formation is altered by loss of Alzheimer’s disease-associated genes. Proc. Natl Acad. Sci. USA https://doi.org/10.1073/pnas.2112095118 (2021). source
    • cites [14] Qi, G. et al. ApoE4 impairs neuron-astrocyte coupling of fatty acid metabolism. Cell Rep. 34, 108572 (2021). source
    • cites [15] Zhao, X., Zhang, S., Sanders, A. R. & Duan, J. Brain lipids and lipid droplet dysregulation in Alzheimer’s disease and neuropsychiatric disorders. Complex Psychiatry 9, 154–171 (2023). source

  • Classified as probably not a substantial claim.

    Premisecitation
    § 100 bp

    MG phagocytosis is important in AD.

    “Considering the importance of MG phagocytosis in AD29,30”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [29] Romero-Molina, C., Garretti, F., Andrews, S. J., Marcora, E. & Goate, A. M. Microglial efferocytosis: diving into the Alzheimer’s disease gene pool. Neuron 110, 3513–3533 (2022). source
    • cites [30] Salter, M. W. & Stevens, B. Microglia emerge as central players in brain disease. Nat. Med. 23, 1018–1027 (2017). source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Introduction

    GWAS analyses have identified 75 genome-wide significant loci for LOAD.

    “GWAS analyses have identified 75 genome-wide significant (P < 5 × 10−8) loci for LOAD1–3.”
    Quote in the paperNot a substantial claim3 of 3 sources checked
    • cites [1] Lambert, J. C. et al. Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer’s disease. Nat. Genet. 45, 1452–1458 (2013). source

      The Abstract supports that GWAS identified genome-wide significant LOAD loci using P < 5 × 10−8. It reports 19 such loci in the combined analysis, in addition to APOE, not 75; this source does not support the cited total. (LLM, conf=0.90)

      Experimental — accuracy not yet measured
    • cites [2] Jansen, I. E. et al. Genome-wide meta-analysis identifies new loci and functional pathways influencing Alzheimer’s disease risk. Nat. Genet. 51, 404–413 (2019). source

      The Results section (Genome-wide meta-analysis for AD status; page unavailable) reports genome-wide significant associations in 29 distinct loci, supporting the general point that GWAS identified loci associated with Alzheimer’s disease. It does not support the attributed total of 75 loci; the sentence also cites other sources, which are not assessed here. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
    • cites [3] Bellenguez, C. et al. New insights into the genetic etiology of Alzheimer’s disease and related dementias. Nat. Genet. 54, 412–436 (2022). source

      The Discussion reports 75 independent loci for Alzheimer’s disease and related dementias (ADD), and the GWAS meta-analysis specifies a genome-wide significance threshold of P ≤ 5 × 10−8. The paper does not establish that all 75 loci are specifically for late-onset Alzheimer’s disease (LOAD); its analysis also includes proxy cases of dementia. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation
    § 0 45 90 135 180

    Filipin only stains free (unesterified) cholesterol, and cellular cholesterol accumulation is associated with excess LD.

    “Although filipin only stains free (unesterified) cholesterol, cellular cholesterol accumulation is associated with excess LD10,37,38.”
    Quote in the paperSubstantial claim3 of 3 sources checked
    • cites [10] Marschallinger, J. et al. Lipid-droplet-accumulating microglia represent a dysfunctional and proinflammatory state in the aging brain. Nat. Neurosci. 23, 194–208 (2020). source

      The Introduction describes lipid droplets as storing lipids that can include cholesterol, but does not establish that cellular cholesterol accumulation is associated with excess lipid droplets. The Discussion (‘What causes lipid droplet formation in LDAM?’; page unavailable) reports low cholesteryl ester in these droplets and suggests demyelination does not contribute to their formation. A search for filipin and free or unesterified cholesterol found no passage addressing the staining claim. (LLM, conf=0.74)

      Experimental — accuracy not yet measured
    • cites [37] Olzmann, J. A. & Carvalho, P. Dynamics and functions of lipid droplets. Nat. Rev. Mol. Cell Biol. 20, 137–155 (2019). source

      The ‘Neutral lipid synthesis and lens formation’ section says cholesterol can be esterified into a lipid-droplet component and that sterol esters can trigger droplet formation. This supports a connection between cholesterol storage and lipid droplets, but it does not establish that cellular cholesterol accumulation is associated with excess droplets. Our copy also does not address what filipin stains. (LLM, conf=0.77)

      Experimental — accuracy not yet measured
    • cites [38] Tcw, J. et al. Cholesterol and matrisome pathways dysregulated in astrocytes and microglia. Cell 185, 2213–2233 (2022). source

      The Results section ('Impaired cholesterol trafficking in APOE4 astrocytes and microglia'; page unavailable) identifies filipin as a measure of intracellular free cholesterol, supporting the first part of the claim. The Discussion mentions lipid-droplet accumulation only while describing another study's finding about choline deficiency and unsaturated fatty acids; it does not establish an association between cellular cholesterol accumulation and excess lipid droplets. (LLM, conf=0.87)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § 0 45 90 135 180

    Cholesterol esters, derived from cellular cholesterol, are often stored in LDs along with triacylglycerols (TGs).

    “Cholesterol esters, derived from cellular cholesterol, are often stored in LDs along with triacylglycerols (TGs)37.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [37] Olzmann, J. A. & Carvalho, P. Dynamics and functions of lipid droplets. Nat. Rev. Mol. Cell Biol. 20, 137–155 (2019). source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion

    PICALM's role in lipid/cholesterol metabolism has been reported in Drosophila glia and HEK293 cells.

    “its role in lipid/cholesterol metabolism has been reported in Drosophila glia and HEK293 cells13,36”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [13] Moulton, M. J. et al. Neuronal ROS-induced glial lipid droplet formation is altered by loss of Alzheimer’s disease-associated genes. Proc. Natl Acad. Sci. USA https://doi.org/10.1073/pnas.2112095118 (2021). source
    • cites [36] Mercer, J. L. et al. Modulation of PICALM levels perturbs cellular cholesterol homeostasis. PLoS ONE 10, e0129776 (2015). source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § 0 45 90 135 180

    PICALM has been shown to be involved in lipid/cholesterol metabolism in Drosophila glia and HEK293 cells.

    “Although PICALM has been shown to be involved in lipid/cholesterol metabolism in Drosophila glia and HEK293 cells13,36”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [13] Moulton, M. J. et al. Neuronal ROS-induced glial lipid droplet formation is altered by loss of Alzheimer’s disease-associated genes. Proc. Natl Acad. Sci. USA https://doi.org/10.1073/pnas.2112095118 (2021). source
    • cites [36] Mercer, J. L. et al. Modulation of PICALM levels perturbs cellular cholesterol homeostasis. PLoS ONE 10, e0129776 (2015). source

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation
    § 0 2 4 6

    MCOLN1 is a ROS sensor in lysosomes regulating autophagy.

    “MCOLN1 is a ROS sensor in lysosomes regulating autophagy49”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [49] Zhang, X. et al. MCOLN1 is a ROS sensor in lysosomes that regulates autophagy. Nat. Commun. 7, 12109 (2016). source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion

    There is a reported genetic interaction between the PICALM risk-allele and APOE4.

    “Given the reported genetic interaction between the PICALM risk-allele and APOE419,58, it will be interesting to empirically examine possible epistasis between these two AD risk factors.”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [19] Haney, M. S. et al. APOE4/4 is linked to damaging lipid droplets in Alzheimer’s disease microglia. Nature 628, 154–161 (2024). source

      The section on ACSL1 and triglycerides after fAβ challenge reports APOE4-dependent lipid-droplet accumulation, not a genetic interaction with PICALM. A search of our copy for “PICALM” found no passage addressing that interaction. (LLM, conf=0.75)

      Experimental — accuracy not yet measured
    • cites [58] Jun, G. et al. Meta-analysis confirms CR1, CLU, and PICALM as alzheimer disease risk loci and reveals interactions with APOE genotypes. Arch. Neurol. 67, 1473–1484 (2010). source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation
    § Methods

    None of the donors had large copy-number variants (>100 kb).

    “All donors were also analysed for copy-number variants, and none had large copy-number variants (>100 kb)61”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [61] Levinson, D. F. et al. Copy number variants in schizophrenia: confirmation of five previous findings and new evidence for 3q29 microdeletions and VIPR2 duplications. Am. J. Psychiatry 168, 302–316 (2011). source

      The Methods (GWAS Assay and Detection of CNVs; Analysis of CNV Association) describe testing samples for CNVs, including variants larger than 100 kb. The Results instead report large CNVs in the paper’s own subjects; the paper does not establish that none of the citing study’s donors had them. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation
    § Discussion

    The APOE4 allele was recently found to induce LD accumulation in MGs and impair neuronal function.

    “given that the APOE4 allele was recently found to induce LD accumulation in MGs and impair neuronal function18,19”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [18] Victor, M. B. et al. Lipid accumulation induced by APOE4 impairs microglial surveillance of neuronal-network activity. Cell Stem Cell 29, 1197–1212 (2022). source
    • cites [19] Haney, M. S. et al. APOE4/4 is linked to damaging lipid droplets in Alzheimer’s disease microglia. Nature 628, 154–161 (2024). source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation
    § Methods

    WASP calibration was performed on raw BAM files to eliminate allelic bias towards reference alleles during alignment.

    “To further eliminate allelic bias towards reference alleles during the alignment step, we performed WASP (v.0.3.4) calibration on the generated raw BAM files68.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [68] van de Geijn, B., McVicker, G., Gilad, Y. & Pritchard, J. K. WASP: allele-specific software for robust molecular quantitative trait locus discovery. Nat. Methods 12, 1061–1063 (2015). source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion

    MG phagocytosis in vivo may also be detrimental through excessive synaptic pruning.

    “MG phagocytosis in vivo may also be detrimental through excessive synaptic pruning57.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [57] Hong, S. et al. Complement and microglia mediate early synapse loss in Alzheimer mouse models. Science 352, 712–716 (2016). source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation
    § Methods

    eSNP-karyotyping was performed for all cell lines used to eliminate potential chromosomal abnormalities.

    “eSNP-karyotyping was performed for all cell lines used to eliminate potential chromosomal abnormalities, as previously described20,21.”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [20] Zhang, S. et al. Allele-specific open chromatin in human iPSC neurons elucidates functional disease variants. Science 369, 561–565 (2020). source

      The paper’s passage describing the 20 iPSC lines and their assays (section labeled Abstract, page unavailable) does not say they underwent eSNP-karyotyping. Searches for eSNP-karyotyping and chromosomal-abnormality screening found only a reference to a separate chromosomal-aberration assay, not a statement that this procedure was performed on all cell lines. (LLM, conf=0.72)

      Experimental — accuracy not yet measured
    • cites [21] Zhang, S. et al. Multiple genes in a single GWAS risk locus synergistically mediate aberrant synaptic development and function in human neurons. Cell Genom. 3, 100399 (2023). source

      The STAR★Methods section on experimental models describes checks of pluripotency and CRISPR off-target editing, but does not say that eSNP-karyotyping was performed on all cell lines or used to screen for chromosomal abnormalities. Searches for eSNP-karyotyping, karyotyping, and chromosomal abnormalities found no passage addressing that procedure. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationcitation
    § Discussion

    PICALM expression in patients with AD may also be influenced by cell type, MG state and disease stage.

    “Although our study indicates a reduced PICALM expression in LOAD-risk-allele iMGs, PICALM expression in patients with AD may also be influenced by cell type, MG state and disease stage31,56.”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [31] Ando, K. et al. PICALM and Alzheimer’s disease: an update and perspectives. Cells https://doi.org/10.3390/cells11243994 (2022). source

      Section 3.3.1 says PICALM expression is dysregulated during AD progression, while section 3.4 and Discussion section 4.1 describe differing expression across cell types in AD brains. Section 3.4 also reports microglial PICALM upregulation correlated with AD pathology, but the cited paper does not establish that microglial state itself influences PICALM expression. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [56] Baig, S. et al. Distribution and expression of picalm in Alzheimer disease. J. Neuropathol. Exp. Neurol. 69, 1071–1077 (2010). source

      The Results and Discussion report that picalm is concentrated in endothelial cells, with relatively little in glial cells, supporting a cell-type distinction. The Discussion also speculates that changes in expression may relate to the stage of disease spread. The paper does not address microglial (MG) state, so that part of the claim is unsupported. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Interpretationown data
    § Discussion

    Many LDs in iMGs are peroxidated, suggesting that ROS may also occur first and the peroxidized lipids are sequestered in LDs.

    “However, many LDs in iMGs were also peroxidated (Extended Data Fig. 11b–e), suggesting that ROS may also occur first and the peroxidized lipids are sequestered in LDs16,42,52.”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites [16] Liu, L., MacKenzie, K. R., Putluri, N., Maletić-Savatić, M. & Bellen, H. J. The glia-neuron lactate shuttle and elevated ROS promote lipid synthesis in neurons and lipid droplet accumulation in Glia via APOE/D. Cell Metab. 26, 719–737 (2017). source
    • cites [42] Liu, L. et al. Glial lipid droplets and ROS induced by mitochondrial defects promote neurodegeneration. Cell 160, 177–190 (2015). source

      The Results sections on increased ROS and lipid peroxidation report that ROS may promote glial lipid-droplet accumulation and that accumulated lipids are peroxidated. They do not establish the proposed sequence in which lipids are peroxidated first and then sequestered in droplets. The observation about iMGs is the citing authors’ own result, not a finding attributed to this source. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
    • cites [52] Sies, H. & Jones, D. P. Reactive oxygen species (ROS) as pleiotropic physiological signalling agents. Nat. Rev. Mol. Cell Biol. 21, 363–383 (2020). source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductcitation
    § Methods

    All ROSMAP participants enrolled without known dementia and agreed to detailed clinical evaluation and brain donation at death.

    “All ROSMAP partici- pants enrolled without known dementia and agreed to detailed clinical evaluation and brain donation at death.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [62] Bennett, D. A. et al. Religious Orders Study and Rush Memory and Aging Project. J. Alzheimers Dis. 64, S161–S189 (2018). source

  • May combine two claims. The quote only partly states this claim.

    Findingown data
    § 1 2 3 4 5 6 7

    The LOAD-risk allele of rs10792832 impairs the uptake of amyloid beta (Aβ) and myelin debris in microglia.

    “However, PICALM-KO cells exhibited reduced (about 50%) phagocytosis of myelin-pHrodo compared with the wild-type (WT) cells (Fig. 3f,g)”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The CRISPR-edited iPS cell lines differ from the non-risk lines only at rs10792832 and not at off-target sites.

  • May combine two claims.

    Findingown data
    § 100 bp

    The LOAD-risk allele of rs10792832 reduces PICALM expression by about 50% in iMGs but not in iAst cells.

    “We found that the risk allele reduced PICALM expression by about 50% (Fig. 2b,c and Extended Data Fig. 3e) in iMGs but not in iAst cells (Extended Data Fig. 3f).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The CRISPR-edited iPS cell lines differ from the non-risk lines only at rs10792832 and not at off-target sites.

  • May combine two claims.

    Findingown data
    § Mapping functional LOAD-risk variants

    The authors confirmed cell-type-specific OCRs of MG (for example, SPI1) and Ast (for example, VIM) cells, and clear separation of iMG and iAst cells from neuronal samples.

    “We confirmed cell-type-specific OCRs of MG (for example, SPI1) and Ast (for example, VIM) cells, and clear separation of iMG and iAst cells from neuronal samples (Extended Data Fig. 1f–i).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data
    § Mapping functional LOAD-risk variants

    The authors identified between 15,693 and 72,291 ASoC SNPs, a large proportion of which were cell type specific.

    “We identified between 15,693 and 72,291 ASoC SNPs, a large proportion of which were cell type specific (Extended Data Fig. 2a, Supplementary Tables 2–6 and Supplementary Fig. 3a).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data
    § 1 2 3 4 5 6 7

    Loss of PICALM did not perturb the Golgi apparatus (Giantin), the distribution of early endosomes (EEA1), AP-4 containing secretory and endocytic vesicles or the internalization of transferrin and cholera toxin through endocytosis.

    “The loss of PICALM did not perturb the Golgi apparatus (Giantin), the distribution of early endosomes (EEA1), AP-4 containing secretory and endocytic vesicles or the internalization of transferrin and cholera toxin through endocytosis (Supplementary Fig. 6).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data
    § 0 2 4 6

    Among the 55 lipid species altered in risk-allele iMGs versus non-risk (FDR < 0.05), 32 were TGs (5.6-fold enrichment, Fisher's exact test, P < 2.6 × 10−16) and all were increased.

    “Among the 55 lipid species altered in risk-allele iMGs (versus non-risk) (FDR < 0.05), 32 were TGs (5.6-fold enrichment, Fisher's exact test, P < 2.6 × 10−16) and all were increased (Extended Data Fig. 10a,b and Supplementary Table 17).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The CRISPR-edited iPS cell lines differ from the non-risk lines only at rs10792832 and not at off-target sites.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data
    § 77. Nunez, J. K. et al. Genome-wide programmable transcriptional memory by CRISPR-based

    ATAC-seq fragment size distribution plots show periodical nucleosome-free region patterns that are similar between samples and different iPSC-derived cell types.

    “(c) ATAC-seq fragment size distribution plots show periodical nucleosome-free region patterns that are similar between samples and different iPSC-derived cell types.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data
    § 77. Nunez, J. K. et al. Genome-wide programmable transcriptional memory by CRISPR-based

    iMG-specific OCR peaks were found for the known MG-specific gene SPI1 (PU.1), and VIM peaks were much stronger in iAs than in other cell types.

    “(f) iMG-specific OCR peaks for a known MG-specific gene SPI1 (also known as PU.1) and much stronger peaks for VIM in iAs than in other cell types.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data
    § 77. Nunez, J. K. et al. Genome-wide programmable transcriptional memory by CRISPR-based

    Cytochalasin D (10 μM) nearly completely inhibits phagocytosis.

    “Cytochalasin D (10 µM) nearly completely inhibits phagocytosis.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data
    § 77. Nunez, J. K. et al. Genome-wide programmable transcriptional memory by CRISPR-based

    iMG carrying the LOAD risk-allele show increased lipid droplet and CellRox density and area compared to non-risk iMG.

    “Quantification of (e) shows increased LD and CellRox density and area in iMG carrying LOAD risk-allele (vs. non-risk).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data
    § 77. Nunez, J. K. et al. Genome-wide programmable transcriptional memory by CRISPR-based

    For genomic PCR products of the same heterozygous samples, the two alleles of rs10792832 have equal Sanger sequencing peak heights.

    “Note the equal peak height of the two alleles for genomic PCR products (sequenced by using primers for on-target editing confirmation) of the same heterozygous samples as in the top panels.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data
    § 77. Nunez, J. K. et al. Genome-wide programmable transcriptional memory by CRISPR-based

    iMG with PICALM-CRISPRoff show increased LD accumulation vs non-risk allele, mimicking the effect of the PICALM risk allele.

    “Note iMG with PICALM-CRISPRoff also show increased LD accumulation (vs. non-risk allele), mimicking the effect of the PICALM risk allele.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data
    § 77. Nunez, J. K. et al. Genome-wide programmable transcriptional memory by CRISPR-based

    iMG carrying the LOAD risk-allele show increased LD area, density per cell, size, and fluorescence intensity vs non-risk for both CD04 and CD09.

    “(b) to (e) Increased LD (BODIPY+) area, density per cell, size, and fluorescence intensity in iMG carrying the LOAD risk-allele (vs. non-risk) for both CD04 and CD09 lines.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisenone given
    § Introduction

    Association-based mapping approaches do not directly implicate functional or causal variants.

    “However, such association-based mapping approaches do not directly implicate functional or causal variants.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion

    PICALM is known to be expressed in brain MGs.

    “While PICALM is known to be expressed in brain MGs51, and its role in lipid/cholesterol metabolism has been reported in Drosophila glia and HEK293 cells13,36, our work demonstrates a human-MG-specific role for PICALM in regulating LD accumulation and phagocytosis, which potentially contribute to LOAD pathophysiology.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [51] Ando, K. et al. Clathrin adaptor CALM/PICALM is associated with neurofibrillary tangles and is cleaved in Alzheimer’s brains. Acta Neuropathol. 125, 861–878 (2013). source

  • Classified as probably not a substantial claim.

    Premisecitation
    § 0 2 4 6

    PICALM has been shown to promote neuron-astrocyte lipid transfer in Drosophila and rat astrocytes.

    “PICALM has been shown to promote neuron–astrocyte lipid transfer in Drosophila and rat astrocytes13.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [13] Moulton, M. J. et al. Neuronal ROS-induced glial lipid droplet formation is altered by loss of Alzheimer’s disease-associated genes. Proc. Natl Acad. Sci. USA https://doi.org/10.1073/pnas.2112095118 (2021). source

  • May combine two claims.

    Interpretationown data
    § 0 2 4 6

    These results indicate cell-type-specific effects of PICALM on lipid transfer and LD formation in MG and Ast cells, warranting further studies.

    “These results indicate cell-type-specific effects of PICALM on lipid transfer and LD formation in two very different types of glial cells, MG and Ast cells, warranting further studies.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The CRISPR-edited iPS cell lines differ from the non-risk lines only at rs10792832 and not at off-target sites.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § 0 2 4 6

    GRN encodes the lysosomal protein progranulin related to neurodegenerative diseases.

    “GRN encodes the lysosomal protein progranulin related to neurodegenerative diseases50.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [50] Simon, M. J., Logan, T., DeVos, S. L. & Di Paolo, G. Lysosomal functions of progranulin and implications for treatment of frontotemporal dementia. Trends Cell Biol. 33, 324–339 (2023). source

  • May combine two claims.

    Interpretationown data
    § Discussion

    The study mechanistically links the putative LOAD causal SNP at the PICALM locus to its MG-specific effect on PICALM expression, resulting in LD accumulation in iMGs partially through lysosomal dysregulation.

    “Moreover, we mechanistically link the putative LOAD causal SNP at the PICALM locus to its MG-specific effect on PICALM expression, resulting in LD accumulation in iMGs partially through lysosomal dysregulation (Extended Data Fig. 11f).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion

    In Drosophila and rat astrocytes, PICALM is required for storing peroxidated lipids sourced from neurons.

    “In Drosophila and rat astrocytes, PICALM is required for storing peroxidated lipids sourced from neurons13.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [13] Moulton, M. J. et al. Neuronal ROS-induced glial lipid droplet formation is altered by loss of Alzheimer’s disease-associated genes. Proc. Natl Acad. Sci. USA https://doi.org/10.1073/pnas.2112095118 (2021). source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § Methods

    The integrated density of pHrodo-myelin within each cell was measured from five random fields of view per biological replicate.

    “within each cell was measured from five random fields of view”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Methods

    All lipids were normalized to the peak area of an internal standard with the same head group as the target lipids.

    “All lipids were normalized to the peak area of an internal standard with the same”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § Methods

    Lipidomics data were acquired in two technical replicates for each sample.

    “with lipidomics data acquired in two technical”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § Methods

    All analysed ATAC-seq samples passed standard quality control based on the characteristic nucleosomal periodicity of the insert fragment size distribution and high signal-to-noise ratio around transcription start sites (TSSs).

    “All analysed ATAC–seq samples passed standard quality control based on the characteristic nucleosomal periodicity of the insert fragment size distribution and high signal-to-noise ratio around transcription start sites (TSSs).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Methods

    The samples from two different iMG differentiations were analysed.

    “The samples from two different iMG differentiations”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Methods

    For comparisons involving more than two groups, possible random effects from experimental rounds were accounted for when applicable.

    “while accounting for any possible random effects from experimental rounds when applicable.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § Methods

    Pluripotency of CRISPR-Cas9-edited human iPS cell lines was confirmed by staining with pluripotency marker antibodies OCT4, NANOG and SSEA4.

    “To confirm the pluripotency of CRISPR–Cas9-edited human iPS cell lines, the cells were stained with pluripotency marker antibodies: rabbit anti-OCT4 (ab181557, Abcam, 1:250), goat anti-NANOG (AF1997-SP, R&D Systems, 1:50) and mouse anti-SSEA4 (ab16287, Abcam, 1:250).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion

    In glia, the main sources of ROS include peroxisomes and mitochondria that perform lipid β-oxidation.

    “In glia, the main sources of ROS include peroxisomes and mitochondria that perform lipid β-oxidation52.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [52] Sies, H. & Jones, D. P. Reactive oxygen species (ROS) as pleiotropic physiological signalling agents. Nat. Rev. Mol. Cell Biol. 21, 363–383 (2020). source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § Methods

    Benjamini-Hochberg correction was applied to all qualified ASoC SNPs in the ATAC-seq analysis of ASoC SNPs.

    “the Benjamini–Hochberg correction was applied”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). The quote only partly states this claim.

    Study design and conductnone given
    § Methods

    Five random fields of view per biological replicate were acquired for quantification of BODIPY+ droplets.

    “Five random fields of view (per biological replicate) were acquired for”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data
    § Discussion

    PICALM has a human-MG-specific role in regulating LD accumulation and phagocytosis, which potentially contribute to LOAD pathophysiology.

    “our work demonstrates a human-MG-specific role for PICALM in regulating LD accumulation and phagocytosis, which potentially contribute to LOAD pathophysiology.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Methods

    Cell line ID (CD04 and CD09) was considered as a coefficient in generalized linear models to further remove line-specific effects.

    “When constructing the generalized linear models, we considered cell line ID (CD04 and CD09) as a coefficient to further remove line-specific effects.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § Methods

    All cell cultures used in the study were confirmed to be mycoplasma-free using a PCR detection kit.

    “All cell cul- tures were confirmed to be mycoplasma-free using the PCR detection kit (ab289834, Abcam).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning
    § Discussion

    The lipids are peroxidated and sequestered in LDs, which are initially protective against ROS elevation.

    “(2) the lipids are peroxidated and sequestered in LDs, which are initially protective against ROS elevation;”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning
    § Methods

    Only SNPs with corresponding rs# records found in dbSNP v154 were retained to reduce bias introduced by acquired or de novo mutations during cell growth.

    “To reduce bias introduced by any acquired (or de novo) mutations during cell growth, only SNPs with corresponding rs# records found in dbSNP v154 were retained.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductreasoning
    § Methods

    The human iPS cell lines used for ATAC-seq were generated using the Sendai virus method and are integration-free.

    “The human iPS cell lines were generated using the Sendai virus method to ensure that they are integration-free”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning
    § Methods

    For two-group comparisons, an LMM was used to account for potential random effects from different experimental rounds and clones.

    “otherwise, we used the R packages lme4 and lmerTest to fit data into an LMM”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning
    § Discussion

    Ongoing lipid build-up causes cellular stress, further elevating ROS and LD levels.

    “(3) ongoing lipid build-up causes cellular stress, further elevating ROS and LD levels;”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Methods

    The median values from each biological replicate were used for statistical analysis of myelin phagocytosis.

    “the median values from each replicate were”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given
    § Methods

    The purity of selected CRISPR-Cas9-edited iPS cell clones was confirmed for on-target editing and the absence of off-target editing.

    “The purity of the selected clones was confirmed for on-target editing and the absence of off-target editing (see below).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Methods

    For quantification of PICALM in IBA1+ microglia, all processing was conducted by a researcher blinded to individual diagnoses.

    “For quantification of PICALM in IBA1+ microglia, all processing was conducted by a researcher blinded to individual diagnoses.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § Methods

    Sanger sequencing was used to verify editing in resistant colonies.

    “Resistant colonies were collected 14 days after transfection and a small amount of DNA from each colony was used for Sanger sequencing to verify editing.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § Methods

    No obvious mapping bias to reference alleles was confirmed by visualizing volcano plots of allelic read-depth ratios against −log2[P] values.

    “We confirmed there was no obvious mapping bias to reference alleles by visualizing the volcano plots that graph the allelic read-depth ratios against −log2[P] values in scatter plots.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Methods

    Primers were designed to amplify regions corresponding to the four top-ranking predicted off-targets to check on-target and off-target editing.

    “Primers were designed to amplify regions corresponding to the four top-ranking predicted off-targets to check on-target and off-target editing.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). The quote only partly states this claim.

    Study design and conductnone given
    § Methods

    The median values from each biological replicate were used for statistical analysis of LD staining.

    “and the median values from each biological replicate”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § Methods

    Benjamini-Hochberg procedure was used to adjust P values accounting for multiple testing in RNA-seq DEG analysis.

    “the Benjamini–Hochberg procedure was used to adjust P values”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Study design and conductnone given
    § Discussion

    We cannot formally exclude a possible long-range cis-effect of the AD-risk variant on genes other than PICALM.

    “However, we cannot formally exclude a possible long-range cis-effect of the AD-risk variant on genes other than PICALM.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Methods

    The images were acquired in a manner that the operator was blinded to the sample and genotype identification.

    “The images were acquired in a manner that the operator was”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given
    § Methods

    ComBat-seq was applied to remove batch effects between the two experiments (rounds of iMG differentiation).

    “The ComBat-seq function from the R package sva was applied to remove batch effects between the two experiments (that is, rounds of iMG differentiation).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data
    § Discussion

    Partial loss of PICALM in MGs exacerbates ROS levels, leading to increased lipid/cholesterol synthesis due to increased SREBP.

    “We therefore propose a model that connects the PICALM risk allele to LD accumulation and phagocytosis deficits (Extended Data Fig. 11f): (1) partial loss of PICALM in MGs exacerbates ROS levels (Extended Data Fig. 7e), leading to increased lipid/cholesterol synthesis due to increased SREBP;”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data
    § 77. Nunez, J. K. et al. Genome-wide programmable transcriptional memory by CRISPR-based

    PICALM knockdown or knockout causes LD accumulation.

    “Extended Data Fig. 9 | PICALM knockdown (KD) or knockout (KO) causes LD accumulation.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data
    § 77. Nunez, J. K. et al. Genome-wide programmable transcriptional memory by CRISPR-based

    The LOAD GWAS SNP risk allele is linked to reduced PICALM expression, lysosomal dysfunction, LD accumulation, LD peroxidation, cellular ROX level, and phagocytosis deficits in iMG.

    “(f) Mechanistic insight on the link between the LOAD GWAS SNP risk allele and the reduced PICALM expression, lysosomal dysfunction, LD accumulation, LD peroxidation, cellular ROX level, and phagocytosis deficits in iMG.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § 77. Nunez, J. K. et al. Genome-wide programmable transcriptional memory by CRISPR-based

    Selected DE genes were validated by qPCR and IF staining.

    “Validation of the selected DE genes by qPCR and IF staining.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § 77. Nunez, J. K. et al. Genome-wide programmable transcriptional memory by CRISPR-based

    PU.1 antibody was validated in PMP cells from which iMG were further differentiated.

    “PU.1 antibody validation in PMP cells from which iMG were further differentiated.”
    Quote in the paperNot a substantial claim
77 of 77

  • The paper relies on this without support.

    The functional link between rs10792832 and PICALM was unknown before this study.

    “and the PICALM locus with an unknown functional link between rs10792832 and its target gene PICALM (Fig. 1c,d and Extended Data Fig. 2l).”

  • The paper relies on this without support.

    The role of PICALM in LD formation in microglia and its link to the LOAD-risk allele was not demonstrated before this study.

    “its potential role in LD formation in MGs, especially its link to the LOAD-risk allele, has not been demonstrated.”

  • The paper relies on this without support.

    The CRISPR-edited iPS cell lines differ from the non-risk lines only at rs10792832 and not at off-target sites.

    “iMGs from a second clone of each iPS cell line also exhibited similar allelic difference in phagocytic activity”

  • The paper relies on this without support.

    Around 70% of ASoC SNPs are in distal enhancers and are likely to be functional.

    “Given that around 70% of ASoC SNPs are in distal enhancers and are likely to be functional, our study presents a framework for prioritizing functional non-coding risk variants through combinatorial analyses of whether a SNP is predicted to alter TF-binding sites and/or is a brain eQTL.”

  • The authors state this as an assumption.

    A functional common GWAS SNP does not display monoallelic chromatin accessibility.

    “with the assumption that a functional common GWAS SNP does not display monoallelic chromatin accessibility”

  • The paper relies on this without support.

    rs10792832 allele G is a LOAD risk allele for Alzheimer's disease.

    “The LOAD risk-allele G of rs10792832 shows reduced PU.1 binding in qPCR of ChIP products.”

  • The paper relies on this without support.

    C11-BODIPY fluorescence shift from red to green indicates lipid peroxidation/ROS.

    “(a) The schematics show C11-BODIPY (581/591), a fluorescent lipid peroxidation sensor, shifts its fluorescence from red to green in the presence of ROS.”

  • The paper relies on this without support.

    The LOAD risk-allele of PICALM does not affect its expression in iAst.

    “over the genetic background of non-risk allele since the LOAD risk-allele of PICALM does not affect its expression in iAst”
8 of 8

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Cytochalasin D concentration differs by 1,000-fold for C20 assay

    Major

    The Fig. 3 caption gives 10 μM cytochalasin D for the C20-cell phagocytosis control, while the C20-cell assay method gives 10 nM.

1 of 1

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Kozlova, A., Zhang, S., Sudwarts, A., Zhang, H., Smirnou, S., Byeon, S. K., Thapa, C., Sun, X., Stephenson, K., Zhao, X., Jamison, B., Ponnusamy, M., He, X., Schneider, J. A., Pandey, A., Bennett, D. A., Pang, Z. P., Sanders, A. R., Bellen, H. J., Thinakaran, G., & Duan, J.
Affiliations
  • Endeavor Health (US)
  • University of South Florida (US)
  • Mayo Clinic (US)
  • University of Chicago (US)
  • Rush University Medical Center (US)
  • Rutgers, The State University of New Jersey (US)
  • Baylor College of Medicine (US)
Corresponding author
  • Thinakaran, Gopal
  • Duan, Jubao
ORCID
Venue
Nature
Publisher
Springer Science and Business Media LLC
Published
2025-09-03
Volume
646
Issue
8087
Pages
1178-1186
ISSN
  • 0028-0836 (ISSN-L)
  • 1476-4687
DOI
10.1038/s41586-025-09486-x
PMC
PMC11142308
OpenAlex
W4413946395
Semantic Scholar
9ecf72980121fc7b9b3948e5c2d369f93c859949
Cited by
53
References
85
Citations by year
2026: 38 · 2025: 12 · 2024: 3
FWCI
14.82
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature (publishedVersion, cc-by-nc-nd)
  • Research Square (acceptedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (submittedVersion, cc-by)
  • Europe PMC (PubMed Central) (submittedVersion, cc-by-nc-nd)
  • Digital Commons-TMC (Texas Medical Center) (submittedVersion)
Funders
  • BrightFocus Foundation
  • University of Miami
  • National Institutes of Health (5r01ag015819-21, R01 AG063175, R01MH106575, R01MH116281)
  • Goizueta Business School, Emory University
  • National Institute on Aging (P30 AG072975, U01 AG072439, R01AG063175, U01AG061356, R01 AG081374, R01 AG017917, U01AG046152, P30AG010161, RF1AG079141, R01AG073260, R01AG079141, R01-AG015819)
  • National Institute of Mental Health (R01MH116281, R01 MH106575)
  • National Cancer Institute (U01CA271410, P30CA015083)
Topics
  • Life Sciences › Neuroscience › Neurology › Neuroinflammation and Neurodegeneration Mechanisms
  • Health Sciences › Medicine › Physiology › Alzheimer's disease research and treatments
  • Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
Keywords
PICALM, Alzheimer's disease, microglia, lipid droplets, amyloid-beta uptake, cholesterol biosynthesis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-16.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-16.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review