Cited by 202Neuroinflammation and Neurodegeneration MechanismsAlzheimer's disease research and treatmentsNeurological Disease Mechanisms and Treatments

APOE4 impairs the microglial response in Alzheimer’s disease by inducing TGFβ-mediated checkpoints

Open access
Yin, Zhuoran, Rosenzweig, Neta, Kleemann, Kilian L., Zhang, Xiaoming, Brandão, Wesley, Margeta, Milica A., Schroeder, Caitlin, Sivanathan, Kisha N. +35 more·2023·Nature Immunology

Abstract

The APOE4 allele is the strongest genetic risk factor for late-onset Alzheimer's disease (AD). The contribution of microglial APOE4 to AD pathogenesis is unknown, although APOE has the most enriched gene expression in neurodegenerative microglia (MGnD). Here, we show in mice and humans a negative role of microglial APOE4 in the induction of the MGnD response to neurodegeneration. Deletion of microglial APOE4 restores the MGnD phenotype associated with neuroprotection in P301S tau transgenic mice and decreases pathology in APP/PS1 mice. MGnD-astrocyte cross-talk associated with β-amyloid (Aβ) plaque encapsulation and clearance are mediated via LGALS3 signaling following microglial APOE4 deletion. In the brains of AD donors carrying the APOE4 allele, we found a sex-dependent reciprocal induction of AD risk factors associated with suppression of MGnD genes in females, including LGALS3, compared to individuals homozygous for the APOE3 allele. Mechanistically, APOE4-mediated induction of ITGB8-transforming growth factor-β (TGFβ) signaling impairs the MGnD response via upregulation of microglial homeostatic checkpoints, including Inpp5d, in mice. Deletion of Inpp5d in microglia restores MGnD-astrocyte cross-talk and facilitates plaque clearance in APP/PS1 mice. We identify the microglial APOE4-ITGB8-TGFβ pathway as a negative regulator of microglial response to AD pathology, and restoring the MGnD phenotype via blocking ITGB8-TGFβ signaling provides a promising therapeutic intervention for AD.

AI snapshot

Written from the paper’s full text.

TL;DR

Microglial APOE4 cell-autonomously suppresses the neurodegenerative microglia (MGnD) response to Alzheimer’s disease pathology via ITGB8–TGFβ signaling and homeostatic checkpoints such as INPP5D. Deleting microglial APOE4, deleting Inpp5d, or blocking ITGB8 restores MGnD/astrocyte cross-talk and reduces pathology in mouse AD models.

Key findings

  • APOE4-expressing microglia showed upregulated homeostatic genes including *Tgfbr2*, *Inpp5d*, *Spi1* and *Smad3* versus APOE3 microglia (RNA-seq, n = 11–14 mice/group), and after apoptotic-neuron injection APOE4 microglia failed to induce MGnD genes such as *Clec7a*, *Itgax*, *Lilr4b*, *Lpl*, *Lgals3* and *Spp1*; phagocytic FCRLS+CD11b+ microglia were reduced (n = 7 APOE3-KI, n = 9 APOE4-KI).
  • Conditional deletion of microglial APOE4 restored MGnD signatures and was neuroprotective in P301S tau mice, with reduced neuronal loss in cortical layer 5 (n = 7–10 mice/group), and in APP/PS1 mice increased Clec7a^hi Lgals3^hi MGnD, reduced HJ3.4B+ plaque pathology and reduced LAMP1+ dystrophic neurites.
  • Mechanistically, APOE4 induced ITGB8–TGFβ signaling; cortical *Itgb8* deletion or anti-ITGB8 neutralizing antibody induced MGnD genes and reduced Aβ plaque size in APP/PS1 mice (n = 8 IgG-treated, n = 9 anti-ITGB8-treated), while anti-ITGB8 restored CLEC7A and GFAP in APP/PS1:APOE4-KI mice (n = 11 ROIs from 6–7 mice/group). Microglial *Inpp5d* deletion also induced MGnD and reduced plaque load.
  • In human AD brains, APOE3/APOE4 heterozygosity showed sex-dependent effects: females had reduced MGnD/astrocyte genes including *LGALS3*, *GFAP*, *VIM* and *S100A* and increased AD risk genes including *ABCA7*, *IFNAR1*, *APP*, *BIN1*, *CD33*, *HAVCR2* and *SMAD3*; pSMAD3 was increased in IBA1+ cells (n = 27 APOE3/3 cells, n = 33 APOE3/4 cells), and LGALS3 immunoreactivity was reduced (n = 8 APOE3/3, n = 9 APOE3/4 donors). Males showed increased *ITGB8*.

Methods

  • Mouse genetics: APOE3-KI and APOE4-KI mice, conditional microglial deletion via Cx3cr1-CreERT2, and crosses to APP/PS1 and P301S tau models; stereotactic injections of apoptotic neurons, recombinant LGALS3, anti-ITGB8 antibody, PU.1 inhibitor, and adoptive transfer of phagocytic microglia.
  • Human tissue: postmortem AD brain samples from Netherlands Brain Bank and Massachusetts Alzheimer’s Disease Research Center; RNA-seq donors included 12 AD APOE3/3 and 13 AD APOE3/4 cases, with additional IHC and RNAscope cohorts.
  • Main assays/analyses: FACS, bulk RNA-seq/Smart-seq2, scRNA-seq, H3K9ac ChIP-seq, lipidomics, immunohistochemistry, RNAscope, RT-qPCR; DESeq2 likelihood-ratio tests, Seurat clustering, IPA/NicheNet/CellTalkDB cross-talk analysis, t-tests and one-way ANOVA.

Limitations

  • The paper has no dedicated limitations section; it states that Cx3cr1-CreERT2 may also affect border-associated macrophages, albeit to a lesser extent, and cautions that prior LGALS3 literature used global *Lgals3*-KO or pre-cultured Aβ models, so microglial-derived LGALS3 conclusions should be interpreted carefully.
  • Evident from the text: human analyses are cross-sectional postmortem comparisons with modest sex-stratified donor numbers (e.g., RNA-seq male n = 5–7, female n = 6–7; IHC n = 8–9), limiting causal and sex-specific inference.
  • Evident from the text: most mechanistic evidence comes from transgenic mouse models and intracranial injections, so translation to human AD progression and therapeutic dosing remains untested.

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

Integrity signals

Registry

Crossref · 10.1038/s41590-023-01627-6

Cites

90 of 91 references verified0 citation errors · 1 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

60 of 83 cited sources checked · 35 supported · 1 flagged · accuracy not yet measured

  • Premisecitation
    § A list of authors and their affiliations appears at the end of the paper · p. 2

    The APOE4 allele is a major genetic risk factor for late-onset Alzheimer's disease.

    “The APOE4 allele is the strongest genetic risk factor for late-onset Alzheimer's disease (AD).”
    • Also stated in: § A list of authors and their affiliations appears at the end of the paper · p. 2 · plain
      “The APOE4 allele is a major genetic risk factor for late-onset AD and was previously shown to accelerate AD progression in humans and in mouse models 6-11 .”
    Quote in the paperSubstantial claim3 of 6 sources checked
    • cites [6] Farrer, L. A. et al. Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. JAMA 278, 1349–1356 (1997). source
    • cites [7] Corder, E. H. et al. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science 261, 921–923 (1993). source
    • cites [8] Strittmatter, W. J. et al. Apolipoprotein E: high-avidity binding to β-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc. Natl Acad. Sci. USA 90, 1977–1981 (1993). source

      The allele-association passage (labeled Materials and Methods; page not provided) reports a significantly higher APOE-e4 frequency in familial Alzheimer disease patients than in controls, supporting its importance as a genetic risk factor. The paper does not compare APOE-e4 with other genetic risk factors to establish the citing sentence’s claim that it is the strongest, and its findings concern predominantly late-onset familial disease rather than all late-onset Alzheimer’s disease. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites [9] Raber, J. et al. Isoform-specific effects of human apolipoprotein E on brain function revealed in ApoE knockout mice: increased susceptibility of females. Proc. Natl Acad. Sci. USA 95, 10914–10919 (1998). source
    • cites [10] Youmans, K. L. et al. APOE4-specific changes in Aβ accumulation in a new transgenic mouse model of Alzheimer disease. J. Biol. Chem. 287, 41774–41786 (2012). source
    • cites [11] Liu, D. S. et al. APOE4 enhances age-dependent decline in cognitive function by down-regulating an NMDA receptor pathway in EFAD-Tg mice. Mol. Neurodegener. 10, 7 (2015). source
  • Premisecitation
    § A list of authors and their affiliations appears at the end of the paper · p. 2

    Neurodegenerative microglia (MGnD), also referred to as disease-associated microglia, are regulated by reciprocal suppression of TGFβ and induction of APOE signaling in different neurodegenerative models, including AD.

    “We identified that neurodegenerative microglia (MGnD) 4 , also referred to as disease-associated microglia 5 , are regulated by the reciprocal suppression of transforming growth factor-β (TGFβ) and induction of apolipoprotein E (APOE) signaling in different neurodegenerative models, including AD 4 .”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [4] Krasemann, S. et al. The TREM2–APOE pathway drives the transcriptional phenotype of dysfunctional microglia in neurodegenerative diseases. Immunity 47, 566–581 (2017). source
    • cites [5] Keren-Shaul, H. et al. A unique microglia type associated with restricting development of Alzheimer’s disease. Cell 169, 1276–1290 (2017). source
  • Premisecitation
    § A list of authors and their affiliations appears at the end of the paper · p. 2

    Microglia gene expression patterns are strongly correlated with late-onset Alzheimer's disease.

    “Genome -wide association meta -analysis studies (GWASs) 1,2 and interactome investigation 3 identified microglia gene expression patterns as strongly correlated with late-onset Alzheimer's disease (AD).”
    Quote in the paperSubstantial claim3 of 3 sources checked
    • cites [1] 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 (page unavailable) reports associations between Alzheimer’s disease risk genes and microglia-specific expression in both mouse and human datasets. But the associations did not survive the stringent Bonferroni correction, and the paper does not report a strong correlation between microglial expression patterns themselves and late-onset Alzheimer’s disease. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
    • cites [2] Wightman, D. P. et al. A genome-wide association study with 1,126,563 individuals identifies new risk loci for Alzheimer’s disease. Nat. Genet. 53, 1276–1282 (2021). source
    • cites [3] Nott, A. et al. Brain cell type-specific enhancer–promoter interactome maps and disease-risk association. Science 366, 1134–1139 (2019). source

      The indexed passage on AD genetics and the microglia interactome (page unavailable) links AD-risk variants to microglia-specific enhancers and identifies 25 microglial genes linked to those variants. This supports an association between microglial gene regulation and AD risk, but it does not report that microglial gene-expression patterns are strongly correlated with late-onset AD. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
  • Premisecitation
    § A list of authors and their affiliations appears at the end of the paper · p. 2

    In the central nervous system, APOE is upregulated in MGnD.

    “In the central nervous system, APOE is upregulated in MGnD 4,5 .”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [4] Krasemann, S. et al. The TREM2–APOE pathway drives the transcriptional phenotype of dysfunctional microglia in neurodegenerative diseases. Immunity 47, 566–581 (2017). source
    • cites [5] Keren-Shaul, H. et al. A unique microglia type associated with restricting development of Alzheimer’s disease. Cell 169, 1276–1290 (2017). source

      The Results report that Apoe expression increases as microglia transition from a homeostatic state to disease-associated microglia (DAM) in an Alzheimer’s mouse model (page unavailable). The paper calls this population DAM, not MGnD, and does not establish in our copy that the two labels are interchangeable. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
  • Premisecitation
    § Deletion of microglial APOE4 restores protective MGnD · p. 4

    In previous studies, brain regions of P301S: APOE4-KI mice demonstrate marked neurodegeneration compared to P301S: APOE3-KI mice at 9.5 months of age.

    “In previous studies, brain regions of P301S: APOE4 -KI mice demonstrate marked neurodegeneration compared to P301S: APOE3 -KI mice at 9.5 months of age, although the MGnD phenotype was similar between the two groups 12,21 .”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [12] Wang, C. et al. Selective removal of astrocytic APOE4 strongly protects against tau-mediated neurodegeneration and decreases synaptic phagocytosis by microglia. Neuron 109, 1657–1674 (2021). source
    • cites [21] Shi, Y. et al. ApoE4 markedly exacerbates tau-mediated neurodegeneration in a mouse model of tauopathy. Nature 549, 523–527 (2017). source

      The paper reports significantly more brain atrophy in P301S/E4 than P301S/E3 mice, particularly in the hippocampus, piriform/entorhinal cortex, and amygdala. The reported comparison is at 9 months, not the claimed 9.5 months (main text, page unavailable). (LLM, conf=0.88)

      Experimental — accuracy not yet measured
  • May combine two claims. 28
  • Classified as probably not a substantial claim. 40
  • The quote only partly states this claim. 3
  • May combine two claims.

    Interpretationcitation
    § Discussion · p. 14

    APOE4 converges to impair the MGnD response to neurodegeneration, which plays a protective role in glaucoma and is detrimental in AD.

    “Thus, APOE4 converges to impair the MGnD response to neurodegeneration, which plays a protective role in glaucoma 14 and is detrimental in AD.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [14] Margeta, M. A. et al. Apolipoprotein E4 impairs the response of neurodegenerative retinal microglia and prevents neuronal loss in glaucoma. Immunity 55, 1627–1644 (2022). source

      The source supports the claim that APOE4 dampens the MGnD response, but says the opposite about that response’s role in glaucoma. Its Results section (“MGnD microglia directly induce RGC degeneration in vivo”; page not available) reports that MGnD microglia caused retinal ganglion cell loss. The Abstract says impaired activation of APOE4 microglia—not the MGnD response—is protective in glaucoma. (LLM, conf=0.90)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § A list of authors and their affiliations appears at the end of the paper · p. 2

    APOE4 was previously shown to accelerate AD progression in humans and in mouse models.

    “The APOE4 allele is a major genetic risk factor for late-onset AD and was previously shown to accelerate AD progression in humans and in mouse models 6-11 .”
    Quote in the paperNot a substantial claim3 of 6 sources checked
    • cites [6] Farrer, L. A. et al. Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. JAMA 278, 1349–1356 (1997). source
    • cites [7] Corder, E. H. et al. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science 261, 921–923 (1993). source
    • cites [8] Strittmatter, W. J. et al. Apolipoprotein E: high-avidity binding to β-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc. Natl Acad. Sci. USA 90, 1977–1981 (1993). source

      The allele-association passage reports a significantly higher APOE-e4 frequency in people with familial Alzheimer disease than in controls, supporting the genetic-risk portion of the citing sentence. It does not show that APOE4 accelerates disease progression in humans or mouse models; searches for progression and mouse models found no such result. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites [9] Raber, J. et al. Isoform-specific effects of human apolipoprotein E on brain function revealed in ApoE knockout mice: increased susceptibility of females. Proc. Natl Acad. Sci. USA 95, 10914–10919 (1998). source
    • cites [10] Youmans, K. L. et al. APOE4-specific changes in Aβ accumulation in a new transgenic mouse model of Alzheimer disease. J. Biol. Chem. 287, 41774–41786 (2012). source

      The Discussion reports accelerated Aβ accumulation in E4FAD mice, supporting the mouse-model portion of the claim. The Abstract reports greater plaque deposition with APOE4 in humans and transgenic mice and identifies APOE4 as a major AD risk factor, but it does not establish accelerated clinical AD progression in humans. The paper's evidence is chiefly about amyloid pathology, not the full course of AD. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [11] Liu, D. S. et al. APOE4 enhances age-dependent decline in cognitive function by down-regulating an NMDA receptor pathway in EFAD-Tg mice. Mol. Neurodegener. 10, 7 (2015). source

      The Introduction reports earlier onset of cognitive deficits in human APOE4 carriers, but says evidence for a faster rate of cognitive decline is conflicting. The Discussion reports a more rapid early decline in spatial recognition memory in E4FAD mice. Thus, the source supports the mouse finding and earlier human onset, but not an unqualified claim that APOE4 accelerates AD progression in humans. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation
    § Results · p. 3

    Cx3cr1 CreERT2-mediated deletion of microglial genes may also affect BAMs, albeit to a lesser extent.

    “Of note, using Cx3cr1 CreERT2 mice to delete microglial genes may also affect border-associated macrophages (BAMs) 16 , albeit to a lesser extent as BAMs were shown to be continuously replenished from skull bone marrow, unlike microglia 17 .”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [16] Goldmann, T. et al. Origin, fate and dynamics of macrophages at central nervous system interfaces. Nat. Immunol. 17, 797–805 (2016). source
    • cites [17] Cugurra, A. et al. Skull and vertebral bone marrow are myeloid cell reservoirs for the meninges and CNS parenchyma. Science 373, eabf7844 (2021). source

      The Discussion (page not available) says skull and vertebral marrow continuously replenish monocytes and neutrophils at CNS borders and that marrow-derived monocytes can become meningeal macrophages. The Introduction describes microglia as self-maintained. The paper does not establish that BAMs as a group are continuously replenished, or that Cx3cr1 CreERT2-mediated deletion affects them to a lesser extent; the latter part of the citing sentence is the portion attached to this citation. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § A list of authors and their affiliations appears at the end of the paper · p. 2

    Deletion of APOE4 in astrocytes markedly reduced neurodegeneration in a mouse model of tauopathy.

    “A recent report showed that deletion of APOE4 in astrocytes markedly reduced neurodegeneration in a mouse model of tauopathy 12 .”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [12] Wang, C. et al. Selective removal of astrocytic APOE4 strongly protects against tau-mediated neurodegeneration and decreases synaptic phagocytosis by microglia. Neuron 109, 1657–1674 (2021). source
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § APOE4 impairs the microglial response via PU.1 · p. 4

    Low expression of PU.1 was suggested by a recent GWAS to be protective against AD.

    “Importantly, low expression of PU.1 was suggested by a recent GWAS to be protective against AD 19 .”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [19] Huang, K. L. et al. A common haplotype lowers PU.1 expression in myeloid cells and delays onset of Alzheimer’s disease. Nat. Neurosci. 20, 1052–1061 (2017). source
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Deletion of microglial APOE4 restores protective MGnD · p. 4

    Neurodegeneration is a pathological hallmark of AD.

    “Neurodegeneration is a pathological hallmark of AD 20 , which is exacerbated in a mouse model of tauopathy expressing APOE4 (refs. 12,21).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [20] Serrano-Pozo, A., Frosch, M. P., Masliah, E. & Hyman, B. T. Neuropathological alterations in Alzheimer disease. Cold Spring Harb. Perspect. Med. 1, a006189 (2011). source
  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation
    § Impaired MGnD and astrocyte activation in individuals carrying the APOE4 allele · p. 9

    HAVCR2 encodes the checkpoint molecule TIM3, a promising therapeutic approach in cancer, and was recently described in a GWAS as an AD risk gene.

    “the latter of which encodes the checkpoint molecule TIM3, one of the most promising new therapeutic approaches in cancer 32 that was recently described in a GWAS as an AD risk gene 2 .”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [32] Wolf, Y., Anderson, A. C. & Kuchroo, V. K. TIM3 comes of age as an inhibitory receptor. Nat. Rev. Immunol. 20, 173–185 (2020). source
    • cites [2] Wightman, D. P. et al. A genome-wide association study with 1,126,563 individuals identifies new risk loci for Alzheimer’s disease. Nat. Genet. 53, 1276–1282 (2021). source
  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation
    § MGnD activate astrocytic plaque encapsulation via LGALS3 · p. 7

    Integrin αvβ8, composed of Itgb8 and Itgav subunits, plays a critical role in TGFβ activation and microglia homeostatic maturation.

    “subunits of integrin α v β8 (ref. 30; Fig. 6b,c and Supplementary Table 5), which plays a critical role in TGFβ activation and microglia homeostatic maturation 31”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [30] Munger, J. S. & Sheppard, D. Cross talk among TGF-β signaling pathways, integrins, and the extracellular matrix. Cold Spring Harb. Perspect. Biol. 3, a005017 (2011). source

      The section “TGF-b Activation by RGD-Binding Integrins” (page unavailable) says that αvβ8 activates TGF-β1 and TGF-β3. Our copy does not address microglial homeostatic maturation, so it supports the activation claim but not that additional role. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
    • cites [31] Arnold, T. D. et al. Impaired αVβ8 and TGFβ signaling lead to microglial dysmaturation and neuromotor dysfunction. J. Exp. Med. 216, 900–915 (2019). source
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion · p. 14

    Higher PU.1 expression in humans is likely to result in earlier AD onset.

    “These results support previous reports demonstrating that higher PU.1 expression in humans is likely to result in earlier AD onset 19,48 .”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [19] Huang, K. L. et al. A common haplotype lowers PU.1 expression in myeloid cells and delays onset of Alzheimer’s disease. Nat. Neurosci. 20, 1052–1061 (2017). source
    • cites [48] Pimenova, A. A. et al. Alzheimer’s-associated PU.1 expression levels regulate microglial inflammatory response. Neurobiol. Dis. 148, 105217 (2021). source

      The Results section reports a human-associated SPI1 haplotype linked to lower PU.1 expression and delayed AD onset, which supports the claimed direction of association. It does not report that higher PU.1 expression was shown to cause earlier onset in humans; this paper’s own experiments used mouse microglial cells. No page number is available in our copy. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion · p. 14

    APOE4 impairs the response of neurodegenerative retinal microglia via LGALS3 signaling associated with neuroprotection in glaucoma.

    “We recently demonstrated that APOE4 impairs the response of neurodegenerative retinal microglia via LGALS3 signaling associated with neuroprotection in glaucoma 14 .”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [14] Margeta, M. A. et al. Apolipoprotein E4 impairs the response of neurodegenerative retinal microglia and prevents neuronal loss in glaucoma. Immunity 55, 1627–1644 (2022). source
  • May combine two claims.

    Premisecitation
    § Discussion · p. 14

    Depletion of reactive astrocytes in AD mice results in increased plaque load, synaptic dysfunction and memory loss.

    “Moreover, reactive astrocytes play a beneficial role in limiting AD pathology, and their depletion in AD mice results in increased plaque load, synaptic dysfunction and memory loss 51 .”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [51] Katsouri, L. et al. Ablation of reactive astrocytes exacerbates disease pathology in a model of Alzheimer’s disease. Glia 68, 1017–1030 (2020). source

      The Results sections on synaptic density and memory report synaptic loss and worsened spatial-memory performance after ablation of proliferating reactive astrocytes in APP23 mice. But the Results section on amyloid-β (page unavailable) explicitly reports no significant change in amyloid plaque burden measured by Thio-S staining; increased amyloid-β staining is not the same as increased plaque load. (LLM, conf=0.91)

      Experimental — accuracy not yet measured
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion · p. 14

    APOE4 is associated with a decreased risk of glaucoma.

    “Interestingly, APOE4 was shown to be associated with a decreased risk of glaucoma 54-56 .”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites [54] Mabuchi, F. et al. The apolipoprotein E gene polymorphism is associated with open angle glaucoma in the Japanese population. Mol. Vis. 11, 609–612 (2005).
    • cites [55] Lam, C. Y. et al. Association of apolipoprotein E polymorphisms with normal tension glaucoma in a Chinese population. J. Glaucoma 15, 218–222 (2006). source
    • cites [56] Margeta, M. A. et al. Association of APOE with primary open-angle glaucoma suggests a protective effect for APOE ε4. Invest. Ophthalmol. Vis. Sci. 61, 3 (2020). source
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion · p. 14

    TGFβ signaling was suggested to play both beneficial and detrimental roles in AD.

    “Although TGFβ signaling was suggested to play both beneficial and detrimental roles in AD 57 , its overproduction in astrocytes promotes cerebrovascular fibrosis and amyloidosis 58 .”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [57] Tesseur, I. et al. Deficiency in neuronal TGF-β signaling promotes neurodegeneration and Alzheimer’s pathology. J. Clin. Invest. 116, 3060–3069 (2006). source
  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation
    § Discussion · p. 14

    TGFβ1 is implicated in vascular dementia to promote abnormal vascular remodeling and was proposed as a therapeutic target for AD.

    “Moreover, TGFβ1 is implicated in vascular dementia to promote abnormal vascular remodeling and was proposed as a therapeutic target for AD 59 .”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [59] Kandasamy, M. et al. TGF-β signaling: a therapeutic target to reinstate regenerative plasticity in vascular dementia? Aging Dis. 11, 828–850 (2020). source

      The sections on VaD pathogenesis and vascular remodeling link TGF-β1 to vascular degeneration, fibrosis, and abnormal remodeling in vascular dementia. The paper proposes targeting TGF-β signaling to treat vascular dementia, but does not specifically propose it as a therapeutic target for Alzheimer’s disease (pages unavailable in our copy). (LLM, conf=0.87)

      Experimental — accuracy not yet measured
  • May combine two claims.

    Premisecitation
    § Discussion · p. 14

    Deletion of Tgfbr2 or Smad2/Smad3 activates astrocytes.

    “A TGFβ-enriched milieu in the brains of individuals with AD that carry the APOE4 allele suppressed astrocyte activation, consistent with previous reports showing astrocyte suppression in response to the TGFβ ligand 50 , and deletion of Tgfbr2 (ref. 31) or Smad2 / Smad3 activates astrocytes.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [31] Arnold, T. D. et al. Impaired αVβ8 and TGFβ signaling lead to microglial dysmaturation and neuromotor dysfunction. J. Exp. Med. 216, 900–915 (2019). source
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Discussion · p. 14

    LGALS3 was reported to promote Aβ oligomerization and toxicity.

    “Although LGALS3 was reported to promote Aβ oligomerization and toxicity, these studies were performed using either global Lgals3 -KO mice or Aβ monomers precultured with LGALS3 before injection into the brain 52,53 .”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [52] Boza-Serrano, A. et al. Galectin-3, a novel endogenous TREM2 ligand, detrimentally regulates inflammatory response in Alzheimer’s disease. Acta Neuropathol. 138, 251–273 (2019). source

      The Results section reports that galectin-3 accompanied the formation of insoluble Aβ aggregates after Aβ monomers were injected into mouse hippocampi; Aβ alone produced no such deposits. The Discussion also links galectin-3 to harmful Aβ-associated inflammatory responses and reduced cognitive performance in 5xFAD mice. But the paper does not establish that galectin-3 promotes Aβ oligomerization specifically, or directly demonstrate increased toxicity of Aβ oligomers. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
    • cites [53] Tao, C. C. et al. Galectin-3 promotes Aβ oligomerization and Aβ toxicity in a mouse model of Alzheimer’s disease. Cell Death Differ. 27, 192–209 (2020). source
  • May combine two claims.

    Premisecitation
    § Discussion · p. 14

    APOE4 acquires a novel function in the nucleus, where it translocates and directly binds DNA to affect transcription of genes associated with AD.

    “In addition to the classical role of APOE in cholesterol transport, APOE4 acquires a novel function in the nucleus, where it translocates and directly binds DNA to affect the transcription of genes associated with AD 45,46 .”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [45] Rohn, T. T. & Moore, Z. D. Nuclear localization of apolipoprotein E4: a new trick for an old protein. Int. J. Neurol. Neurother. 4, 067 (2017). source
    • cites [46] Theendakara, V. et al. Direct transcriptional effects of apolipoprotein E. J. Neurosci. 36, 685–700 (2016). source
  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductcitation
    § Antibodies · p. 39

    Validation data for non-commercial antibodies (anti-Fcrls, anti-Apoe, anti-Abeta, anti-Itgb8) were published in cited works.

    “Validation data for non-commercial antibodies: anti-Fcrls mab was published by Butovsky et al. 2014 (PMID: 24316888). Validation data for anti-Apoe mab was published by Kim et al. 2012 (PMID: 23129750), anti-Abeta mab was published by Bero et al. 2012 (PMID: 22457485), and anti-Itgb8 mAb was published by Dodagatta-Marri et al., 2021 (PMID: 34233193).”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites Butovsky et al. 2014 source
    • cites Kim et al. 2012

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Bero et al. 2012 source

      The Materials and methods section (Aβ plaque deposition; page unavailable) identifies HJ3.4 as an in-house anti-Aβ monoclonal antibody, and the Results report its use to stain and quantify Aβ plaques. These passages support the antibody’s identification and use, but do not present antibody-specific validation data, such as specificity controls. The other antibodies in the citing sentence are attributed to other works, not to Bero et al. (LLM, conf=0.82)

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

    Study design and conductcitation
    § Antibodies · p. 38

    Validation data for all commercial antibodies are available on the manufacturer's website, including previously published papers.

    “Validation data for all commercial antibodies are available on the manufacturer's website including previously published papers as below:”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites Krasemann et al., 2017 source

      The STAR*METHODS Key Resources Table lists antibody manufacturers and catalog numbers, but does not say that validation data for all commercial antibodies are available on manufacturers’ websites or in previously published papers. Searches for the quoted statement and for antibody validation and manufacturer websites found no passage addressing that availability claim; the indexed passages provide no page number. (LLM, conf=0.72)

      Experimental — accuracy not yet measured
    • cites Lin et al., 2019

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Staszewski et al., 2019

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

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

    Study design and conductcitation
    § Life sciences study design · p. 38

    Sample sizes were justified by prior publications using APP/PS1 and P301S mouse models that provided estimates of expected effect sizes and variance.

    “Sample sizes were chosen based on prior publications using the APP/PS1 and P301S mouse models, which provided estimations of expected effect sizes and variance.”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites Radde et al., 2006 source

      The Cerebral amyloidosis results and Fig. 6 report APPPS1-21 outcomes, sample sizes, and measures of variability that could inform estimates for that model. The paper does not study P301S mice or state that its results were used to choose the citing study’s sample sizes. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites Shi et al., 2017 source

      The Methods > Animals passage says sample sizes were chosen using estimates to provide 80% power to detect differences of 20% or greater. It does not say those estimates came from prior APP/PS1 or P301S publications, or specify estimates of variance. No page number is available in our copy. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
  • The quote only partly states this claim.

    Findingown data
    § APOE4 impairs the microglial response via PU.1 · p. 4

    Immunohistochemistry showed increased astrocytic activation, as determined by Serpina3n expression around Aβ plaques, in APP/PS1: APOE4-KI mice treated with PU.1 inhibitors.

    “Immunohistochemistry showed an enhanced CLEC7A + MGnD response and increased astrocytic activation, as determined by Serpina3n expression around β-amyloid (Aβ) plaques (Fig. 2i-m and Supplementary Table 2).”
    Quote in the paperNot a substantial claim
  • The quote only partly states this claim.

    Findingown data
    § APOE4 impairs the microglial response via PU.1 · p. 4

    Immunohistochemistry showed enhanced CLEC7A+ MGnD response in APP/PS1: APOE4-KI mice treated with PU.1 inhibitors.

    “Immunohistochemistry showed an enhanced CLEC7A + MGnD response and increased astrocytic activation, as determined by Serpina3n expression around β-amyloid (Aβ) plaques (Fig. 2i-m and Supplementary Table 2).”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Findingown data
    § MGnD activate astrocytic plaque encapsulation via LGALS3 · p. 7

    IPA identified IGF-1 as the most activated upstream regulator, whereas TGFβ1 was the top suppressed upstream regulator in astrocytes in response to APOE4-cKO MGnD.

    “IPA identified IGF-1 as the most activated upstream regulator, whereas TGFβ1 was the top suppressed upstream regulator in astrocytes in response to APOE4 -cKO MGnD (Fig. 5m).”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Findingown data
    § Targeting ITGB8 enhances the MGnD response · p. 12

    RNA-seq analysis of cortical microglia revealed an induction of MGnD-related genes (Apoe, Cd300ld, Cd74 and Axl) and suppression of homeostatic genes (Tmem119, Siglech, Mertk and Havcr2).

    “RNA-seq analysis of cortical microglia revealed an induction of MGnD-related genes ( Apoe , Cd300ld , Cd74 and Axl ), while the expression of homeostatic genes ( Tmem119 , Siglech , Mertk and Havcr2 ) was suppressed (Fig. 8e and Supplementary Table 8).”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Findingown data
    § Targeting ITGB8 enhances the MGnD response · p. 11

    Impaired induction of the MGnD signature and astrocyte activation in individuals with AD that carry the APOE4 allele.

    “Fig. 7 | Impaired induction of the MGnD signature and astrocyte activation in individuals with AD that carry the APOE4 allele.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Findingown data
    § Targeting ITGB8 enhances the MGnD response · p. 12

    Reduced SMAD3 phosphorylation and increased APOE immunoreactivity in Itgb8-cKO mice.

    “Moreover, we found reduced SMAD3 phosphorylation and increased APOE immunoreactivity in Itgb8 -cKO mice (Fig. 8g-i).”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Findingown data
    § Targeting ITGB8 enhances the MGnD response · p. 10

    Genetic deletion of Itgb8 specifically in the cortex in Emx1Cre:Itgb8fl/fl (Itgb8-cKO) mice results in a complete absence of homeostatic TMEM119+ microglia and high expression of LGALS3 and CLEC7A in cortical microglia.

    “Genetic deletion of Itgb8 specifically in the cortex in Emx1 Cre : Itgb8 fl/fl ( Itgb8 -cKO) mice showed a complete absence of homeostatic TMEM119 + microglia, while LGALS3 and CLEC7A were highly expressed in cortical microglia (Fig. 8a and Extended Data Fig. 8a-c).”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Findingown data
    § Discussion · p. 14

    LGALS3 is a critical MGnD molecule that is restored in microglia following deletion of APOE4 and is associated with increased astrocytic activation and reduced plaque pathology.

    “We identified LGALS3 as a critical MGnD molecule that is restored in microglia following the deletion of APOE4 and is associated with increased astrocytic activation and reduced plaque pathology.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Findingown data
    § Discussion · p. 14

    Microglial pSMAD3 immunoreactivity, a marker of TGFβ signaling, was increased in the brains of females with AD carrying the APOE4 allele, indicative of a homeostatic microglial signature.

    “Moreover, microglial pSMAD3 immunoreactivity as a marker of TGFβ signaling was increased in the brains of females with AD that carry the APOE4 allele, indicative of a homeostatic microglial signature.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The MGnD signature observed in mouse microglia corresponds to the human MGnD signature.
  • May combine two claims.

    Findingown data
    § Discussion · p. 14

    Genetic deletion of Itgb8 or Inpp5d was sufficient to restore the MGnD response and astrocyte activation and reduce plaque load in AD mice.

    “Mechanistically, the genetic deletion of Itgb8 or Inpp5d was sufficient to restore the MGnD response and astrocyte activation and reduce plaque load in AD mice.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Findingown data
    § Discussion · p. 14

    In brains of individuals with AD carrying the APOE4 allele, there is a sex-dependent impairment of the MGnD signature, with pronounced reduction in astrocyte activation in females.

    “In the brains of individuals with AD that carry the APOE4 allele, we found a sex-dependent impairment of the MGnD signature, with a pronounced reduction in astrocyte activation observed in females.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The MGnD signature observed in mouse microglia corresponds to the human MGnD signature.
  • May combine two claims.

    Interpretationown data
    § APOE4 impairs the microglial response via PU.1 · p. 4

    Targeting microglial APOE4 or PU.1 can restore MGnD induction.

    “These findings indicate that APOE4 predisposes microglia toward a homeostatic signature that interferes with microglial responses to acute neurodegeneration, and targeting microglial APOE4 or PU.1 can restore MGnD induction.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Interpretationown data
    § Deletion of Inpp5d in microglia ameliorates pathology in mice with AD · p. 12

    Blocking ITGB8-TGFβ signaling enhances the MGnD response and reduces AD pathology in APP/PS1 mice.

    “Fig. 8 | Blocking ITGB8-TGFβ signaling enhances the MGnD response and reduces AD pathology in APP/PS1 mice.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Interpretationown data
    § Targeting ITGB8 enhances the MGnD response · p. 12

    The reproducibility in microglial gene expression profiles indicates that regional deletion of Itgb8 suppresses TGFβ signaling, leading to an induction of the MGnD phenotype.

    “The reproducibility in microglial gene expression profiles indicates that regional deletion of Itgb8 suppresses TGFβ signaling, leading to an induction of the MGnD phenotype.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Interpretationown data
    § Deletion of Inpp5d in microglia ameliorates pathology in mice with AD · p. 12

    These results support the role of INPP5D as a microglial homeostatic checkpoint, and its deletion is beneficial in mitigating AD pathology via the induction of the MGnD response.

    “These results support the role of INPP5D as a microglial homeostatic checkpoint, and its deletion is beneficial in mitigating AD pathology via the induction of the MGnD response.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning
    § Immunohistochemistry · p. 20

    Tissues that were not adequately perfused were excluded from further analysis to eliminate autofluorescence associated with blood contamination.

    “Tissues that were not adequately perfused were not further analyzed to eliminate autofluorescence associated with blood contamination.”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Interpretationreasoning
    § Discussion · p. 14

    Nuclear localization of APOE4 may directly repress the transcription of MGnD genes, a function absent in APOE3-expressing microglia, which is alleviated following deletion of APOE4.

    “These results suggest that nuclear localization of APOE4 may directly repress the transcription of MGnD genes, a function absent in APOE3-expressing microglia, which is alleviated following the deletion of APOE4 .”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Bulk RNA-seq · p. 19

    Biological outliers were determined by PCA plotting and heat map visualization and removed for final analysis.

    “Biological outliers were determined by PCA plotting and heat map visualization and removed for final analysis.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Statistical analysis · p. 21

    All comparison groups were assessed for normal distribution.

    “All comparison groups were assessed for normal distribution.”
    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 conductnone given
    § Mice · p. 17

    All experimental procedures using animals were approved by the Institutional Animal Care and Use Committee at Brigham and Women's Hospital, Harvard Medical School and University of California, San Francisco.

    “All the experimental procedures using animals were approved by the Institutional Animal Care and Use Committee at Brigham and Women's Hospital, Harvard Medical School and University of California, San Francisco.”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Interpretationreasoning
    § Discussion · p. 14

    APOE may have a dual role in controlling MGnD responses based on its subcellular localization, which may be altered in APOE4-expressing microglia.

    “Thus, APOE may have a dual role in controlling MGnD responses based on its subcellular localization, which may be altered in APOE4-expressing microglia, accounting for the different impact on neurodegeneration detected in tau mice following the deletion of microglial APOE3 compared to APOE4 .”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing). The quote only partly states this claim.

    Study design and conductreasoning
    § Human brain tissue and neuropathology · p. 17

    The postmortem interval for human brain tissue used for RNA-seq was limited to 10 h to ensure tissue quality.

    “The postmortem interval was limited to 10 h to ensure tissue quality.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Bulk RNA-seq · p. 19

    Potential technical outliers were removed for further analysis.

    “Potential technical outliers were removed for further analysis.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data
    § Human brain tissue and neuropathology · p. 17

    Donors of the human brain tissue used for RNA-seq gave informed consent to donate their brains to the Netherlands Brain Bank.

    “Donors gave informed consent to donate their brains to NBB.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Human brain tissue and neuropathology · p. 17

    The study was approved by the Ethics Committees of the Brigham Women's Hospital.

    “The study was approved by the Ethics Committees of the Brigham Women's Hospital and was compliant with the World Medical Association Declaration of Helsinki on ethical principles for medical research involving humans.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § RT-qPCR · p. 17

    All RT-qPCRs were performed in duplicate.

    “All RT-qPCRs were performed in duplicate, and levels of mRNA are presented as relative expression normalized to the expression of Gapdh (mean ± s.e.m.).”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation
    § Discussion · p. 14

    Overproduction of TGFβ in astrocytes promotes cerebrovascular fibrosis and amyloidosis.

    “Although TGFβ signaling was suggested to play both beneficial and detrimental roles in AD 57 , its overproduction in astrocytes promotes cerebrovascular fibrosis and amyloidosis 58 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [58] Wyss-Coray, T. et al. Amyloidogenic role of cytokine TGF-β1 in transgenic mice and in Alzheimer’s disease. Nature 389, 603–606 (1997). source
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning
    § Human brain tissue and neuropathology · p. 17

    The study was compliant with the World Medical Association Declaration of Helsinki on ethical principles for medical research involving humans.

    “The study was approved by the Ethics Committees of the Brigham Women's Hospital and was compliant with the World Medical Association Declaration of Helsinki on ethical principles for medical research involving humans.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Statistical analysis · p. 21

    Sample sizes for the experiments were chosen based on prior publications using APP/PS1 and P301S mice and human analysis, which defined the expected effect size and variance.

    “Sample sizes for experiments were chosen based on prior publications using APP/PS1 and P301S mice and human analysis, which defined the expectations for the effect size and variance.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Immunohistochemistry · p. 20

    Two negative controls were routinely used in immunostaining procedures: staining with an isotype control antibody followed by the secondary antibody, or staining with the secondary antibody alone.

    “The following two negative controls were routinely used in immunostaining procedures: (1) staining with isotype control antibody followed by the secondary antibody or (2) staining with the secondary antibody alone.”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Premisecitation
    § Discussion · p. 14

    APOE4 reduces the activated microglia signature in response to AD pathology in human brains and in human induced pluripotent stem cell-derived microglia with the APOE4 allele.

    “In line with these findings, Liu et al. demonstrated that APOE4 reduces the activated microglia signature in response to AD pathology in human brains and in human induced pluripotent stem cell-derived microglia with the APOE4 allele.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § scRNA-seq · p. 18

    For scRNA-seq, cells with more than 200 sequencing reads and less than 20% mitochondrial transcripts were selected.

    “Cells with more than 200 sequencing reads and less than 20% mitochondrial transcripts were selected, and genes with more than 2 reads across all samples entered downstream analyses.”
    Quote in the paperNot a substantial claim
  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given
    § Life sciences study design · p. 38

    All data were replicated in at least three biological replicates and independent experiments.

    “All data were replicated in at least three biological replicates and independent experiments.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § Additional information · p. 31

    Itgb8 is expressed in the adult mouse brain.

    “Extended Data Fig. 7 | Expression of Itgb8 in the adult mouse brain.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Zhang et al. 38

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Allen Brain Atlas

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

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

    Study design and conductnone given
    § Methodology · p. 40

    Unmapped and duplicate fragments were discarded.

    “We discarded unmapped and duplicate fragments.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Additional information · p. 31

    Data are representative of 2 independent experiments.

    “Data are representative of 2 independent experiments.”
    • Also stated in: § Extended Data Fig. 8 | Blocking Itgb8 signaling enhances MGnD response · p. 32 · plain
      “Data are representative of 2 independent experiments.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Life sciences study design · p. 38

    Mice were allocated to control or experimental groups randomly.

    “Mice were allocated to control or experimental groups randomly.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Life sciences study design · p. 38

    No samples were excluded from analysis.

    “No samples were excluded from analysis.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Research · p. 40

    All experimental procedures using animals were approved by the Institutional Animal Care and Use Committee at Brigham and Women's Hospital, Harvard Medical School and University of California San Francisco.

    “All experimental procedures using animals were approved by the Institutional Animal Care and Use Committee at Brigham and Women's Hospital, Harvard Medical School and University of California San Francisco.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Life sciences study design · p. 38

    For all experiments, investigators were blinded to genotype for data collection and analysis.

    “For all the experiments, investigators were blinded to genotype for data collection and analysis.”
    Quote in the paperNot a substantial claim
  • Classified as probably not a substantial claim.

    Study design and conductown data
    § Methodology · p. 40

    Gating for Ly6c negative and CD11b positive yielded a cell population that was >95% positive for FCRLS, a microglia-specific marker.

    “Gating for Ly6c negative and CD11b positive yielded a cell population which was >95% positive for FCRLS, a microglia specific marker which gates were drawn for and positive cells taken for analysis.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: FCRLS is a microglia-specific marker.
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § Methodology · p. 40

    Raw fastq files were checked for quality using Multiqc sequence analysis.

    “Raw fastq files were first checked for quality using Multiqc sequence analysis.”
    Quote in the paperNot a substantial claim
62 of 62
  • The paper relies on this without support.

    The contribution of microglial APOE4 to AD pathogenesis is unknown.

    “The contribution of microglial APOE4 to AD pathogenesis is unknown, although APOE has the most enriched gene expression in neurodegenerative microglia (MGnD).”
  • The paper relies on this without support.

    The cell-autonomous role of APOE4 expressed by microglia in the regulation of AD pathology has not been addressed.

    “Several studies investigated the molecular signature of microglia in mice globally expressing APOE4 (refs. 10,11); however, the cell-autonomous role of APOE4 expressed by microglia in the regulation of AD pathology has not been addressed.”
  • The authors state this as an assumption.

    The impact of microglial APOE4 on astrocyte phenotype and function is unknown.

    “The impact of microglial APOE4 on astrocyte phenotype and function is unknown.”
  • The paper relies on this without support.

    Recombinant LGALS3 injected into hippocampus and cortex mimics endogenous LGALS3 signaling in APP/PS1 mice.

    “we treated APP/PS1 mice with two different doses of recombinant LGALS3 protein (rLGALS3) or with PBS by injection into the hippocampus and cortex at the age of 4 months.”
  • The paper relies on this without support.

    Gene expression changes measured in whole human brain RNA-seq reflect changes in microglial/astrocyte gene expression rather than only changes in cell composition.

    “we used RNA-seq analysis of whole human brains. We found a significant reduction in the expression levels of MGnD genes, including CLEC7A , AXL , LYZ , CD300LG and HLA -DQB2 , and an induction in expression of the upstream regulator of TGFβ signaling ITGB8 in the brains of male individuals with AD heterozygous for the APOE3 and APOE4 alleles (Fig. 7a and Supplementary Tables 6 and 7).”
  • The paper relies on this without support.

    Mouse models (APP/PS1, P301S) and APOE4-KI/cKO manipulations recapitulate key aspects of human AD and APOE4 biology.

    “These findings in humans are in line with our previous results demonstrating enriched Spi1 expression in APOE4 -KI mice (Fig. 1a,b) and restoration of the MGnD signature following genetic deletion of Spi1 (Fig. 2a,b and Supplementary Table 2).”
  • The paper relies on this without support.

    Publicly available human scRNA-seq and snRNA-seq datasets are representative and comparable for APOE genotype analyses.

    “Analysis of publicly available human brain scRNA-seq data 34 showed increased expression of TGFβ-signaling genes and downstream AD risk factor genes, including INPP5D , in females that are heterozygous for the APOE3 and APOE4 alleles.”
  • The paper relies on this without support.

    Mouse models of AD (APP/PS1, tau, P301S) faithfully represent key aspects of human AD pathology and microglial responses.

    “To test whether enhanced microglia activation and phagocytosis can reduce pathological hallmarks of AD in adult APP/PS1 mice, we used a monoclonal neutralizing antibody to ITGB8 (ADWA11) 43 .”
  • The paper relies on this without support.

    The MGnD signature observed in mouse microglia corresponds to the human MGnD signature.

    “These findings in mice and in humans suggest that APOE4 predisposes to AD pathology in part via the induction of TGFβ-dependent microglia homeostatic regulators 13 that impair microglia-astrocyte cross-talk in response to neurodegeneration.”
  • The paper relies on this without support.

    FCRLS is a microglia-specific marker.

    “FCRLS, a microglia specific marker”
10 of 10

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?

  • Removing APOE4 specifically from microglia of APOE4/PS19 tauopathy mice rescued a neuroprotective microglial signature.

    “removing APOE4 specifically from microglia of APOE4/PS19 tauopathy mice rescued a neuroprotective microglial signature157,158.”

    From Multifaceted roles of APOE in Alzheimer disease (2024) · cites this paper as [158]

    The section “Deletion of microglial APOE4 restores protective MGnD” (p. 4) reports that deleting APOE4 in microglia restored the protective MGnD signature in P301S tauopathy mice. Our copy calls the mice P301S, not PS19, so it does not establish the claim’s more specific model designation. (LLM, conf=0.78)

    Experimental — accuracy not yet measured
  • Women have increased susceptibility to E4 related phenotypes compared to men.

    “women have increased susceptibility to E4 related phenotypes compared to men (Rajan et al., 2021; Yin et al., 2023)”

    From APOE4 reshapes the lipid droplet proteome and modulates microglial inflammatory responses (2025) · cites this paper as Yin et al., 2023

    The Discussion (p. 14) reports a sex-dependent APOE4-associated impairment, with a pronounced reduction in astrocyte activation in females. The results (p. 9) also report APOE4-associated changes in males. These passages support a female-pronounced cellular phenotype in people with AD, but not the broader claim that women are more susceptible than men to E4-related phenotypes generally. (LLM, conf=0.84)

    Experimental — accuracy not yet measured
  • Previous studies investigating the role of microglial E4 in amyloid and tau models have found that microglial E4 suppresses complement and lysosomal pathways, impairs lipid metabolism, and blunts glial responses to plaques, while its deletion restores a neuroprotective phenotype relative to E3.

    “Previous studies investigating the role of microglial E4 in amyloid and tau models have found that microglial E4 suppresses complement and lysosomal pathways, impairs lipid metabolism, and blunts glial responses to plaques, while its deletion restores a neuroprotective phenotype relative to E3”

    From Older adult APOE4 mice exhibit modest immunometabolic adaptation following microglial APOE2 replacement (2027) · cites this paper as Yin et al., 2023

    Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

    Experimental — accuracy not yet measured
  • Another recent study that knocked out Inpp5d in myeloid cells at 6 weeks of age found reduced plaque load, accompanied by a reduction in dystrophic neurites and an increase in GFAP+ immunoreactivity.

    “However, another recent study that knocked out Inpp5d in myeloid cells at 6 weeks of age found reduced plaque load, accompanied by a reduction in dystrophic neurites and an increase in GFAP+ immunoreactivity [172].”
  • APOE is involved in the regulation of DAMs and the microglial response to amyloid.

    “APOE is involved in the regulation of DAMs and the microglial response to amyloid [57, 64–66].”

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.

  • Reporting summary denies exclusions described in RNA-seq Methods

    Minor

    The reporting summary states that no samples were excluded, but the bulk RNA-seq Methods explicitly say technical and biological outliers were removed from analysis.

  • Claim of at least three biological replicates conflicts with two-mouse groups

    Minor

    The reporting summary says all data had at least three biological replicates, while the scRNA-seq figure legend specifies two mice per group.

2 of 2

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Authors
Yin, Z., Rosenzweig, N., Kleemann, K. L., Zhang, X., Brandão, W., Margeta, M. A., Schroeder, C., Sivanathan, K. N., Silveira, S., Gauthier, C., Mallah, D., Pitts, K. M., Durao, A., Herron, S., Shorey, H., Cheng, Y., Barry, J.-L., Krishnan, R. K., Wakelin, S., Rhee, J., Yung, A., Aronchik, M., Wang, C., Jain, N., Bao, X., Gerrits, E., Brouwer, N., Deik, A., Tenen, D. G., Ikezu, T., Santander, N. G., McKinsey, G. L., Baufeld, C., Sheppard, D., Krasemann, S., Nowarski, R., Eggen, B. J. L., Clish, C., Tanzi, R. E., Madore, C., Arnold, T. D., Holtzman, D. M., & Butovsky, O.
Affiliations
  • Brigham and Women's Hospital (US)
  • Boston University (US)
  • Washington University in St. Louis (US)
  • University Medical Center Groningen (NL)
  • Broad Institute (US)
  • Harvard University (US)
  • University of California, San Francisco (US)
  • Massachusetts General Hospital (US)
Corresponding author
Butovsky, Oleg
Venue
Nature Immunology
Published
2023-09-25
Volume
24
Issue
11
Pages
1839-1853
ISSN
  • 1529-2908 (ISSN-L)
  • 1529-2916
OpenAlex
W4387012769
Cited by
202
References
86
Citations by year
2026: 56 · 2025: 75 · 2024: 61 · 2023: 10
FWCI
16.67
Type
journal article
Language
en
Open access
Locations
  • Nature Immunology (publishedVersion)
  • PubMed (publishedVersion)
  • PubMed Central (submittedVersion)
  • HAL (Le Centre pour la Communication Scientifique Directe) (submittedVersion)
  • Oskar-Bordeaux (Universite de Bordeaux) (submittedVersion)
  • University of Groningen research database (University of Groningen / Centre for Information Technology) (submittedVersion, other-oa)
Funders
  • U.S. Department of Health & Human Services | NIH | National Institute on Aging (R01AG051812, R01AG054672, R01AG075509, R21AG076982, R01AG080992)
  • U.S. Department of Health & Human Services | NIH | National Eye Institute (R01EY027921, K12EY016335, K08 EY030160)
  • U.S. Department of Health & Human Services | NIH | National Institute of Neurological Disorders and Stroke (R01NS088137, R21NS101673, RF1NS090934)
  • BrightFocus Foundation (2020A016806)
  • Cure Alzheimer’s Fund
  • U.S. Department of Health & Human Services | NIH | National Institute of General Medical Sciences (R01GM132668)
  • Nancy Davis Foundation for Multiple Sclerosis
  • Alzheimer’s Association (AARF-21-846786)
  • National Multiple Sclerosis Society (FG-2108-38372)
  • U.S. Department of Defense (MS210257)
  • Research to Prevent Blindness
  • Glaucoma Research Foundation
Topics
  • Life Sciences › Neuroscience › Neurology › Neuroinflammation and Neurodegeneration Mechanisms
  • Health Sciences › Medicine › Physiology › Alzheimer's disease research and treatments
  • Life Sciences › Neuroscience › Neurology › Neurological Disease Mechanisms and Treatments
Keywords
APOE ε4, Alzheimer's disease, microglia, neurodegeneration, TGF-beta signaling, ITGB8, galectin-3, INPP5D, sex differences, APP/PS1 mice, neuroprotection
MeSH terms
Alzheimer Disease, Animals, Disease Models, Animal, Female, Humans, Mice, Transgenic, Amyloid beta-Peptides, Microglia, Galectin 3, Mice, Apolipoprotein E4

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 2026-10-18.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-18.

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