CoChat
Sign Up
Cited by 74Neuroinflammation and Neurodegeneration MechanismsSingle-cell and spatial transcriptomicsMultiple Sclerosis Research Studies

Culture shock: microglial heterogeneity, activation, and disrupted single-cell microglial networks in vitro

Open access
Cadiz, Mika P., Jensen, Tanner D., Sens, Jonathon P., Zhu, Kuixi, Song, Won-Min, Zhang, Bin, Ebbert, Mark, Chang, Rui +1 more·2022·Molecular Neurodegeneration
DOI 10.1186/s13024-022-00531-1PMC PMC8962153
Open this paper in CoChat — free accountOpen PDF

Abstract

Background: Microglia, the resident immune cells of the brain, play a critical role in numerous diseases, but are a minority cell type and difficult to genetically manipulate in vivo with viral vectors and other approaches. Primary cultures allow a more controlled setting to investigate these cells, but morphological and transcriptional changes upon removal from their normal brain environment raise many caveats from in vitro studies. Methods: To investigate whether cultured microglia recapitulate in vivo microglial signatures, we used single-cell RNA sequencing (scRNAseq) to compare microglia freshly isolated from the brain to primary microglial cultures. We performed cell population discovery, differential expression analysis, and gene co-expression module analysis to compare signatures between in vitro and in vivo microglia. We constructed causal predictive network models of transcriptional regulators from the scRNAseq data and identified a set of potential key drivers of the cultured phenotype. To validate this network analysis, we knocked down two of these key drivers, C1qc and Prdx1, in primary cultured microglia and quantified changes in microglial activation markers. Results: We found that, although often assumed to be a relatively homogenous population of cells in culture, in vitro microglia are a highly heterogeneous population consisting of distinct subpopulations of cells with transcriptional profiles reminiscent of macrophages and monocytes, and are marked by transcriptional programs active in neurodegeneration and other disease states. We found that microglia in vitro presented transcriptional activation of a set of “culture shock genes” not found in freshly isolated microglia, characterized by strong upregulation of disease-associated genes including Apoe , Lyz2, and Spp1 , and downregulation of homeostatic microglial markers, including Cx3cr1 , P2ry12 , and Tmem119 . Finally, we found that cultured microglia prominently alter their transcriptional machinery modulated by key drivers from the homeostatic to activated phenotype. Knockdown of one of these drivers, C1qc , resulted in downregulation of microglial activation genes Lpl, Lyz2, and Ccl4. Conclusions: Overall, our data suggest that when removed from their in vivo home environment, microglia suffer a severe case of “culture shock”, drastically modulating their transcriptional regulatory network state from homeostatic to activated through upregulation of modules of culture-specific genes. Consequently, cultured microglia behave as a disparate cell type that does not recapitulate the homeostatic signatures of microglia in vivo. Finally, our predictive network model discovered potential key drivers that may convert activated microglia back to their homeostatic state, allowing for more accurate representation of in vivo states in culture. Knockdown of key driver C1qc partially attenuated microglial activation in vitro, despite C1qc being only weakly upregulated in culture. This suggests that even genes that are not strongly differentially expressed across treatments or preparations may drive downstream transcriptional changes in culture.

AI snapshot

Written from the paper’s full text.

TL;DR

Cultured primary microglia are highly heterogeneous and transcriptionally activated relative to freshly isolated microglia, showing a “culture shock” signature. Causal network modeling identified key drivers, and knockdown of C1qc partially reduced microglial activation markers in vitro.

Key findings

  • In vitro astrocyte-plated microglia were heterogeneous: 4855 myeloid cells formed 7 distinct clusters with markers of macrophages, monocytes, proliferation, and neurodegeneration, while homeostatic markers such as Tmem119, Cx3cr1, and P2ry12 were sparse. Freshly isolated microglia were largely homeostatic: 8794 cells, with clusters 0–3 accounting for over 96%, activated cluster 4 at 2.4%, interferon cluster 5 at 1.4%, and activated/interferon cells totaling 3.8%.
  • Cultured microglia showed a “culture shock” signature versus freshly isolated microglia, with upregulation of disease-associated genes including Apoe, Lyz2, Lpl, and Spp1 and downregulation of homeostatic markers Cx3cr1, P2ry12, and Tmem119. In astro-plated versus freshly isolated pseudo-bulk analysis, 231 genes were significantly increased and 9 decreased at adjusted p < 0.01 and absolute average logFC > 1.
  • Integrated CCA of 13,817 cells (8794 freshly isolated, 4855 astro-plated, 168 micro-isolated) separated in vitro from freshly isolated microglia. Causal networks shared 3847 genes (77%, p < 2.2e-16) and identified 444 key drivers, including C1qc and Prdx1.
  • C1qc knockdown in mixed glial co-culture significantly reduced activation genes Lpl, Lyz2, and Ccl4 (n = 3 mice per group, paired t test, p < 0.05 or p < 0.01), while Prdx1 knockdown did not significantly affect activation genes. Prdx1 was 3.3 times higher in culture, whereas C1qc was only 1.03 times higher.

Methods

  • Single-cell RNA sequencing (10X Genomics Chromium) was performed on microglia freshly isolated from ~4-month-old C57BL/6J female mice and on primary microglia from P1 mice prepared as astrocyte-plated co-culture and micro-isolated monoculture.
  • Analyses included Seurat clustering and differential expression, canonical correlation analysis integration, gene set analysis, MEGENA co-expression network analysis, and predictive causal network modeling with key driver analysis.
  • Validation used siRNA knockdown of C1qc or Prdx1 in mixed glial co-culture followed by RT-qPCR for activation genes, with n = 3 mice per group and two-tailed paired t tests.

Limitations

  • Stated: the monoculture comparison was limited by the comparatively small number of microglia captured during sequencing; C1qc knockdown efficiency was variable and modest; Prdx1 knockdown was robust but did not significantly affect activation genes; validation of network targets remains necessary.
  • Stated: whether a triple neuron-astrocyte co-culture system recapitulates the in vivo phenotype is unknown.
  • Evident from the text: scRNAseq used only two mice per condition, qPCR validation used n = 3 mice per group, only female C57BL/6J mice were studied, cultures were derived from P1 mice while freshly isolated microglia came from 3.5–4-month-old mice, and validation was limited to transcript levels.

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

Integrity signals

Registry

Crossref · 10.1186/s13024-022-00531-1

Cites

34 of 34 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

16 of 27 cited sources checked · 9 supported · 0 flagged · accuracy not yet measured

  • Findingown data

    Most markers of developmental diversity are robustly expressed across all clusters of primary cultured microglia rather than in a cluster-specific fashion, except macrophage cluster (cluster 0) and proliferative cluster (cluster 6).

    “We plotted markers associated with microglia in the developing mouse brain from Hammond et al. 2018 [22] and found that most markers of developmental diversity are robustly expressed across all clusters rather than in a cluster-specific fashion, except macrophage cluster (cluster 0) and proliferative cluster (cluster 6) (Fig. 1D).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [22] Hammond TR, Dufort C, Dissing-Olesen L, Giera S, Young A, Wysoker A, et al. Single-cell RNA sequencing of microglia throughout the mouse lifespan and in the injured brain reveals complex cell-state changes. Immunity. 2019;50:253–271.e6. source
  • Premisecitation

    N9 and BV2 cell lines and primary cultured microglia derived from P1 and P2 mice lack the molecular signatures of adult microglia in vivo, including key homeostatic genes Tmem119, P2ry12, Cx3cr1, and Hexb.

    “However, N9 and BV2, two common microglial cell lines, as well as primary cultured microglia derived from P1 and P2 mice, lack the molecular signatures of adult microglia in vivo, including key homeostatic genes Tmem119, P2ry12, Cx3cr1, and Hexb [8].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [8] Butovsky O, Jedrychowski MP, Moore CS, Cialic R, Lanser AJ, Gabriely G, et al. Identification of a unique TGF-β–dependent molecular and functional signature in microglia. Nat Neurosci. 2014;17:131–43. source

      The Results section, “Microglia cell lines do not express a microglia signature” (page unavailable), reports that N9, BV2, and primary cultured P1–P2 microglia did not express the identified adult microglial signature. The Results section, “Identification of a unique microglial signature” (page unavailable), identifies Tmem119, P2ry12, and Hexb as microglial genes, but the copy does not establish that all four named genes—including Cx3cr1—are lacking in each of those preparations. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
  • Findingown data

    The activated microglia cluster had increased expression of genes upregulated in microglia during late response to neurodegenerative disease.

    “microglia during late response to neurodegenerative disease [31]”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [31] Mathys H, Adaikkan C, Gao F, Young JZ, Manet E, Hemberg M, et al. Temporal tracking of microglia activation in Neurodegeneration at single-cell resolution. Cell Rep. 2017;21:366–80. source
  • Findingcitation

    The activated microglia cluster had increased expression of genes upregulated in disease-associated microglia.

    “the activated microglia cluster had increased expression of genes upregulated in disease-associated microglia [5]”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [5] Keren-Shaul H, Spinrad A, Weiner A, Matcovitch-Natan O, Dvir-Szternfeld R, Ulland TK, et al. A unique microglia type associated with restricting development of Alzheimer’s disease. Cell. 2017;169(7):1276–90. source
  • Premisecitation

    In general, microglia across in vitro systems suffer from disseminated molecular changes attributed to loss of signaling with other cells in the brain microenvironment that is required to maintain them in a homeostatic state.

    “In general, microglia across in vitro systems suffer from disseminated molecular changes attributed to loss of signaling with other cells in the brain microenvironment that is required to maintain them in a homeostatic state.”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [8] Butovsky O, Jedrychowski MP, Moore CS, Cialic R, Lanser AJ, Gabriely G, et al. Identification of a unique TGF-β–dependent molecular and functional signature in microglia. Nat Neurosci. 2014;17:131–43. source

      The Results (‘Microglia cell lines do not express a microglia signature’; page not provided) show that the tested cell lines and primary newborn cultures lacked the adult microglial signature. The Results on TGF-β in vitro also show that adding TGF-β1 and MCSF made cultured adult microglia more closely resemble freshly isolated cells. The paper does not establish that microglia across in vitro systems undergo widespread changes because they lose signaling from other brain cells; that causal explanation and scope go beyond its findings. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites [9] Delbridge ARD, Huh D, Brickelmaier M, Burns JC, Roberts C, Challa R, et al. Organotypic brain slice culture microglia exhibit molecular similarity to acutely-isolated adult microglia and provide a platform to study Neuroinflammation. Front Cell Neurosci. 2020;444. source

      The Results report extensive gene-expression differences in conventionally cultured microglia, and the Discussion links preservation of microglial identity to interactions with other CNS cells (page numbers unavailable). But the paper describes that link as a proposed explanation, not an established cause. It also finds that organotypic brain-slice cultures—an in vitro system—preserve microglial identity better, so the claim is too broad in applying the changes across in vitro systems. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
  • Classified as probably not a substantial claim. 9
  • May combine two claims. 25
  • The quote only partly states this claim. 1

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

    Premisecitation

    Disease-associated microglia (DAM) identified in AD and ALS mouse models and aging mice are characterized by amoeboid morphology, phagocytosis of amyloid plaques, upregulation of Apoe, Trem2, Cst7, and Lpl, and downregulation of Tmem119, Hexb, Cx3cr1, and P2ry12.

    “For instance, disease-associated microglia (DAM), that have been identified in AD and ALS mouse models and aging mice, are characterized by amoeboid morphology, phagocytosis of amyloid plaques, upregulation of disease genes including Apoe, Trem2, Cst7, and Lpl, and downregulation of homeostatic genes including Tmem119, Hexb, Cx3cr1, and P2ry12 [5, 6].”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [5] Keren-Shaul H, Spinrad A, Weiner A, Matcovitch-Natan O, Dvir-Szternfeld R, Ulland TK, et al. A unique microglia type associated with restricting development of Alzheimer’s disease. Cell. 2017;169(7):1276–90. source

      The Results (Figures 1, 2, and 4) identify DAM in AD and ALS mouse models and aged mice, report Aβ phagocytosis, and support the stated changes in Apoe, Trem2, Cst7, Lpl, Tmem119, Cx3cr1, and P2ry12. But the Results (Figure 2D) say Hexb expression does not change during the transition to DAM, rather than decreasing. The cited passages do not establish amoeboid morphology. (LLM, conf=0.90)

      Experimental — accuracy not yet measured
    • cites [6] Kang SS, Ebbert MTW, Baker KE, Cook C, Wang X, Sens JP, et al. Microglial translational profiling reveals a convergent APOE pathway from aging, amyloid, and tau. J Exp Med. 2018;215:2235–45. source

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

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Microglia in organotypic brain slice cultures resemble in vivo microglia more than primary cultured microglia, about 50% are non-homeostatic, and overexpress markers associated with microglia in neurodegenerative disease.

    “Furthermore, while microglia in organotypic brain slice cultures resemble in vivo microglia more than primary cultured microglia, about 50% of these cells are non-homeostatic (either proliferative, inflammatory, or reactive), and overexpress markers associated with microglia in neurodegenerative disease [9].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [9] Delbridge ARD, Huh D, Brickelmaier M, Burns JC, Roberts C, Challa R, et al. Organotypic brain slice culture microglia exhibit molecular similarity to acutely-isolated adult microglia and provide a platform to study Neuroinflammation. Front Cell Neurosci. 2020;444. source

      The Results section reports that slice-culture microglia more closely resemble acutely isolated adult microglia than primary cultured microglia, and identifies proliferative, inflammatory, and reactive clusters. It also reports increased expression of neurodegeneration-associated signatures, particularly at three weeks. Our copy does not establish that about 50% of slice-culture microglia are non-homeostatic; cluster proportions vary with culture timepoint, which the claim does not specify. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

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

    Premisecitation

    The addition of TGF-β to culture aids microglia in developing a transcriptional signature that is somewhat more similar to what is observed in vivo.

    “supporting an earlier finding that the addition of TGF-β to culture aids microglia in developing a transcriptional signature that is somewhat more similar to what is observed in vivo [8]”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [8] Butovsky O, Jedrychowski MP, Moore CS, Cialic R, Lanser AJ, Gabriely G, et al. Identification of a unique TGF-β–dependent molecular and functional signature in microglia. Nat Neurosci. 2014;17:131–43. source

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

    Premisecitation

    Baxter et al. recently demonstrated that co-culture with both neurons and astrocytes reduces microglial activation compared to monoculture, and co-culture with only neurons or only astrocytes.

    “Baxter et al. [35] recently demonstrated that co-culture with both neurons and astrocytes reduces microglial activation compared to monoculture, and co-culture with only neurons or only astrocytes”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [35] Baxter PS, Dando O, Emelianova K, He X, McKay S, Hardingham GE, et al. Microglial identity and inflammatory responses are controlled by the combined effects of neurons and astrocytes. Cell Rep. 2021;34(12):108882. source

      The Results section, “Neurons and astrocytes regulate the microglial response to LPS” (page unavailable), reports that the combined co-culture suppressed microglial responses to low-dose LPS more strongly than monoculture or co-culture with either cell type alone. The claim omits that condition: the paper says combined co-culture did not reduce basal expression of LPS-response genes or their maximal LPS-induced levels. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingown data

    Knockdown of key driver C1qc partially attenuated microglial activation in vitro, despite C1qc being only weakly upregulated in culture.

    “Knockdown of key driver C1qc partially attenuated microglial activation in vitro, despite C1qc being only weakly upregulated in culture.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Freshly isolated microglia were highly homogenous, and consisted predominantly of homeostatic or resting microglia, with a small number of activated or interferon-associated cells.

    “In contrast, freshly isolated microglia were highly homogenous, and consisted predominantly of homeostatic or resting microglia, with a small number of activated or interferon-associated cells”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Freshly isolated microglia from 3.5-4-month-old mice formed a highly homogenous population; clusters 0,1,2,3 accounted for over 96% of total microglia.

    “Freshly isolated cells formed a highly homogenous population; clusters 0, 1, 2, and 3 formed a homogenous subpopulation accounting for over 96% of the total microglia.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The culture shock signature is characterized by strong upregulation of disease-associated genes including Apoe, Lyz2 and Spp1, and downregulation of homeostatic microglial markers, such as Cx3cr1, P2ry12, and Tmem119.

    “This signature is characterized by strong upregulation of disease-associated genes including Apoe, Lyz2 and Spp1, and downregulation of homeostatic microglial markers, such as Cx3cr1, P2ry12, and Tmem119”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    C1qc was the least upregulated (1.03 times higher in culture, with expression in 99.5% of cultured cells and 100% of freshly isolated cells).

    “C1qc was the least upregulated (1.03 times higher in culture, with expression in 99.5% of cultured cells and 100% of freshly isolated cells)”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Interferon-related and LPS-related gene sets were lowly expressed across most subpopulations but most enriched in clusters 3 and 4.

    “Markers upregulated in response to virus (Ifit1, Irf7) and LPS-injection (Ifitm3, Crip1) seen in the interferon related and LPS related gene sets, were both lowly expressed across most subpopulations, but were most enriched in clusters 3 and 4 (Fig. 2C-D).”
    Quote in the paperNot a substantial claim

  • The quote only partly states this claim.

    Findingown data

    Both the activated and interferon clusters had lower expression of the resting microglia markers than the homeostatic cluster.

    “both the activated and interferon clusters had lower expression of the resting microglia markers”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Friedman et al. [4] Friedman BA, Srinivasan K, Ayalon G, Meilandt WJ, Lin H, Huntley MA, et al. Diverse brain myeloid expression profiles reveal distinct microglial activation states and aspects of Alzheimer’s disease not evident in mouse models. Cell Rep. 2018;22:832–47. source

  • May combine two claims.

    Findingown data

    The neurodegeneration gene set was robustly expressed at high levels across all clusters, most strongly in cluster 1.

    “Finally, the neurodegeneration gene set that contained the canonical markers of activated microglia such as Lpl and Cst7 was robustly expressed at high levels across all clusters, though most strongly in cluster 1 (Fig. 2G).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Cultured microglia consist of disparate subtypes, whose vastly different transcriptional profiles suggest a non-uniform response to experimental challenges.

    “our data showed that cultured microglia consist of disparate subtypes, whose vastly different transcriptional profiles suggest a non-uniform response to experimental challenges”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Microglia in vitro are marked by widespread activation and are highly heterogenous, consisting of multiple transcriptionally distinct subpopulations.

    “We found that microglia in vitro are marked by widespread activation and are highly heterogenous, consisting of multiple transcriptionally distinct subpopulations”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The macrophage gene set was most enriched in cluster 0 but also present at lower levels in all other clusters.

    “The macrophage gene set, that included the markers F13a1 and Mrc1, was most enriched in cluster 0 but also present at lower levels in all of the other clusters (Fig. 2B).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Substantial overlap was found in the microglial activation signature under both culture conditions when freshly isolated ex vivo microglia was used as a comparator, with commonalities between micro-isolated monocultures and astro-plated cultures.

    “Substantial overlap was found in the microglial activation signature under both culture conditions when freshly isolated ex vivo microglia was used as a comparator, with commonalities between micro-isolated monocultures and astro-plated cultures”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The remaining clusters showed more branching and distinct separation, segregating by microglial preparation method.

    “the remaining clusters showed more branching and distinct separation, segregating by microglial preparation method”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    When astro-plated cultures were compared to micro-isolated monocultured microglia, many differentially expressed genes were identified, including up-regulation of disease- and activation-associated genes, Apoe, C1qa, and C1qb in astro-plated cultures, and up-regulation of genes such as Fth1 and Hmox1 in micro-plated monocultures.

    “When astro-plated cultures were compared to micro-isolated monocultured microglia, we identified many differentially expressed genes, including up-regulation of disease- and activation-associated genes, Apoe, C1qa, and C1qb in astro-plated cultures, and up-regulation of genes such as Fth1 and Hmox1 in micro-plated monocultures”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Despite the similarities, these microglia still differentially expressed key genes associated with Tmem119highP2ry12high but Lyz2lowApoelow homeostatic markers vs Tmem119lowP2ry12low but Lyz2highApoehigh activation markers.

    “despite the similarities, these microglia still differentially expressed key genes associated with Tmem119highP2ry12high but Lyz2lowApoelow homeostatic markers vs Tmem119lowP2ry12low but Lyz2highApoehigh activation markers”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    More than 90% of genes from most gene sets were expressed in cleaned myeloid in vitro cells, except Monocyte (78.0%) and LPS-related (87.2%).

    “We see that with the exception of the Monocyte and LPS-related gene set—which had 78.0 and 87.2% respectively—more than 90% of genes from most gene sets were expressed”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Compared to freshly isolated microglia, astro-plated microglia and micro-isolated monocultured microglia upregulated activation markers such as Lyz2, Apoe, and Lpl and downregulated homeostatic markers P2ry12, Cx3cr1, Sparc, and Tmem119.

    “Compared to freshly isolated microglia, astro-plated microglia and micro-isolated monocultured microglia upregulated activation markers such as Lyz2, Apoe, and Lpl and downregulated homeostatic markers P2ry12, Cx3cr1, Sparc, and Tmem119”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The UMAP of cluster 4 showed that they were spatially separate and mostly non-overlapping subpopulations.

    “the UMAP of cluster 4 showed that they were spatially separate and mostly non-overlapping subpopulations”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Knockdown of C1qc resulted in significant reduction in expression levels of microglial activation genes Lpl, Lyz2, and Ccl4, while Trem2 and Cst7 showed a trend toward downregulation.

    “we observed significant reduction in expression levels of microglial activation genes Lpl, Lyz2, and Ccl4, while Trem2 and Cst7 showed a trend toward downregulation”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The effect of C1qc knockdown occurred despite C1qc being only weakly differentially expressed between cultured and ex vivo microglia and despite only a modest knockdown of C1qc itself.

    “this effect occurred despite C1qc being only weakly differentially expressed between cultured and ex vivo microglia and despite the fact that we achieved only a modest knockdown of C1qc itself”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    Primary cultured microglia show limited signal of developmental diversity, which is insufficient to explain the extensive heterogeneity in these cultured cells.

    “These data suggest that our primary cultured microglia show limited signal of developmental diversity, which is insufficient to explain the extensive heterogeneity in these cultured cells.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Microglial activation encompasses a spectrum of heterogeneous phenotypes that vary according to disease or insult types.

    “Microglial activation, rather than being a singular homogenous cell state, encompasses a spectrum of heterogeneous phenotypes that vary according to disease or insult types [3, 4].”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [3] Ransohoff RM. A polarizing question: do M1 and M2 microglia exist? Nat Neurosci. 2016;19:987–91. source
    • cites [4] Friedman BA, Srinivasan K, Ayalon G, Meilandt WJ, Lin H, Huntley MA, et al. Diverse brain myeloid expression profiles reveal distinct microglial activation states and aspects of Alzheimer’s disease not evident in mouse models. Cell Rep. 2018;22:832–47. source

  • May combine two claims.

    Interpretationown data

    Cultured microglia are not merely activated in the absence of stimuli, they are overactivated, and the phenotype of microglial activation in culture and microglial activation in vivo are dissimilar.

    “cultured microglia are not merely activated in the absence of stimuli, they are overactivated, and the phenotype of microglial activation in culture and microglial activation in vivo are dissimilar”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Freshly isolated ex vivo microglia are a valid proxy for microglia in the brain (in vivo).

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

    Premisecitation

    Friedman et al. identified a set of distinct gene signatures corresponding to various insults and diseases including neurodegenerative disease, LPS administration, and glioma.

    “In a meta-analysis, Friedman et al. [4] identified a set of distinct gene signatures corresponding to various insults and diseases including neurodegenerative disease, lipopolysaccharide (LPS) administration, and glioma.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [4] Friedman BA, Srinivasan K, Ayalon G, Meilandt WJ, Lin H, Huntley MA, et al. Diverse brain myeloid expression profiles reveal distinct microglial activation states and aspects of Alzheimer’s disease not evident in mouse models. Cell Rep. 2018;22:832–47. source

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

    Premisecitation

    The binding of neuronal CD200 to its microglial receptor CD200R is necessary for restricting microglial activation in disease models including rotenone neurotoxicity and Toxoplasma encephalitis.

    “the binding of neuronal CD200 to its microglial receptor CD200R is necessary for restricting microglial activation in disease models including rotenone neurotoxicity [32] and Toxoplasma encephalitis [33]”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [32] Wang X-J, Zhang S, Yan Z-Q, Zhao Y-X, Zhou H-Y, Wang Y, et al. Impaired CD200–CD200R-mediated microglia silencing enhances midbrain dopaminergic neurodegeneration: roles of aging, superoxide, NADPH oxidase, and p38 MAPK. Free Radic Biol Med. 2011;50:1094–106. source
    • cites [33] Deckert M, Sedgwick JD, Fischer E, Schlüter D. Regulation of microglial cell responses in murine toxoplasma encephalitis by CD200/CD200 receptor interaction. Acta Neuropathol. 2006;111:548–58. source

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    Key driver discovery is important for network modeling.

    “highlighting the importance of key driver discovery for network modeling”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Because microglia require signaling from neighboring cells to maintain their homeostatic phenotype, loss of this resting state and increase in activation is an inevitable problem across all in vitro systems.

    “Because microglia require signaling from neighboring cells to maintain their homeostatic phenotype, loss of this resting state and increase in activation is an inevitable problem across all in vitro systems”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The two in vitro culture methods used are representative of all in vitro microglia systems.

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

    Study design and conductown data

    This analysis is limited by the comparatively small number of microglia from monoculture captured during sequencing.

    “though this analysis is limited by the comparatively small number of microglia from monoculture captured during sequencing”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    Validation of key driver targets remains necessary for parsing out biological significance.

    “validation of these targets remains necessary for parsing out biological significance”
    Quote in the paperNot a substantial claim
33 of 33

  • The paper relies on this without support.

    Freshly isolated microglia from adult mouse cortex accurately represent the in vivo homeostatic microglial transcriptome.

    “To compare the transcriptome of the astrocyte-plated cultured microglia to microglia in vivo, we profiled cells freshly isolated from non-transgenic, 3.5-4-month-old mice.”

  • The paper relies on this without support.

    Primary cultured microglia derived from astrocyte-microglial co-culture and microglial monocultures are representative of in vitro microglia generally.

    “To investigate whether cultured microglia recapitulate in vivo microglial signatures, we used single-cell RNA sequencing (scRNAseq) to compare microglia freshly isolated from the brain to primary microglial cultures.”

  • The paper relies on this without support.

    Primary microglia cultures derived from astrocyte-microglial co-culture and microglial monocultures are one of the most commonly used methods for primary microglial culture.

    “we used single-cell RNA sequencing (scRNAseq) to profile the transcriptomes of microglia freshly isolated from the brain of wild-type C57BL/6 J mice and compared them to scRNAseq data generated from primary microglia cultures which are derived from astrocyte-microglial co-culture and microglial monocultures that have been shaken off the astrocyte co-culture, as this is still one of the most”

  • The paper relies on this without support.

    Freshly isolated ex vivo microglia are a valid proxy for microglia in the brain (in vivo).

    “Taken together, these data indicate that in vitro microglia are over-activated in comparison to microglia in vivo”

  • The paper relies on this without support.

    The two in vitro culture methods used are representative of all in vitro microglia systems.

    “loss of this resting state and increase in activation is an inevitable problem across all in vitro systems”
5 of 5

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.

  • Top key driver is identified as both Rpl32 and Rpl132

    Minor

    The listed top key drivers include Rpl32, but the later list of eight differentially expressed targets from that top ten instead names Rpl132.

  • Interferon-related gene-set detection percentage is incorrect

    Minor

    Table 1 reports that all 52 of 52 interferon-related genes were detected, which is 100%, not 91.2%.

    Reported 91.2 · recomputed 100

2 of 2

Cites

Flagged first, then by citations

Research works printed but not linked: 1

  1. Zhang Bin ZJ. Identification of key causal regulators in gene networks. Lect Notes Eng Comput Sci. 2013;2:1309–12.

Cited by

Sorted by citations

Related

Authors
Cadiz, M. P., Jensen, T. D., Sens, J. P., Zhu, K., Song, W.-M., Zhang, B., Ebbert, M., Chang, R., & Fryer, J. D.
Affiliations
  • Mayo Clinic in Arizona (US)
  • University of Arizona (US)
  • Icahn School of Medicine at Mount Sinai (US)
  • University of Kentucky (US)
  • Mayo Clinic (US)
Corresponding author
  • Chang, Rui
  • Fryer, John D.
ORCID
Venue
Molecular Neurodegeneration
Publisher
Springer Science and Business Media LLC
Published
2022-03-28
Volume
17
Issue
1
Pages
26-26
ISSN
1750-1326 (ISSN-L)
DOI
10.1186/s13024-022-00531-1
PMC
PMC8962153
OpenAlex
W4220862875
Semantic Scholar
d5cd793c4a4025a5a9b1777bbe7962b8a8c055a1
Cited by
73
References
33
Citations by year
2026: 12 · 2025: 18 · 2024: 23 · 2023: 14 · 2022: 6
FWCI
4.33
Type
journal article
Language
en
Open access
Licence
Locations
  • Molecular Neurodegeneration (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • UA Campus Repository (The University of Arizona) (acceptedVersion, cc-by)
  • PubMed Central (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
Funders
  • Mayo Foundation for Medical Education and Research
  • Ben Dov Family Luminescence Foundation
  • Ed and Ethel Moore Alzheimer’s Disease Research Program of Florida Department of Health (6AZ06)
  • Coins for Alzheimer's Research Trust
  • CureAlz Foundation
  • JW and HM Goodman Family Foundation
  • National Institutes of Health (NS084974, AG062556, AG062110, NS094137, AG057997, AG062077, NS110435, AG047327, AG049992, AG057457, AG062620, AG057931, AG059093, AG046170, AG068030)
Topics
  • Life Sciences › Neuroscience › Neurology › Neuroinflammation and Neurodegeneration Mechanisms
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics
  • Health Sciences › Medicine › Pathology and Forensic Medicine › Multiple Sclerosis Research Studies
Keywords
microglia, culture shock, microglial heterogeneity, microglial activation, single-cell transcriptomics, transcriptional regulatory networks, C1QC, peroxiredoxin 1, CX3CR1, apolipoprotein E, disease-associated genes, microglial homeostasis, gene knockdown, in vitro-in vivo discrepancy
MeSH terms
Brain, Homeostasis, Macrophages, Down-Regulation, Microglia

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 2026-10-20.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-20.

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