CoChat
Sign Up
Cited by 124Single-cell and spatial transcriptomicsLiver physiology and pathologyLiver Disease Diagnosis and Treatment

Spatial transcriptomics of healthy and fibrotic human liver at single-cell resolution

Open access
Watson, Brianna R., Paul, Biplab, Rahman, Raza Ur, Amir-Zilberstein, Liat, Segerstolpe, Åsa, Epstein, Eliana T., Murphy, Shane, Geistlinger, Ludwig +6 more·2025·Nature Communications
DOI 10.1038/s41467-024-55325-4PMC PMC11697218

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

Abstract Single-cell RNA sequencing (scRNA-seq) has advanced our understanding of cell types and their heterogeneity within the human liver, but the spatial organization at single-cell resolution has not yet been described. Here we apply multiplexed error robust fluorescent in situ hybridization (MERFISH) to map the zonal distribution of hepatocytes, resolve subsets of macrophage and mesenchymal populations, and investigate the relationship between hepatocyte ploidy and gene expression within the healthy human liver. We next integrated spatial information from MERFISH with the more complete transcriptome produced by single-nucleus RNA sequencing (snRNA-seq), revealing zonally enriched receptor-ligand interactions. Finally, analysis of fibrotic liver samples identified two hepatocyte populations that are not restricted to zonal distribution and expand with injury. Together these spatial maps of the healthy and fibrotic liver provide a deeper understanding of the cellular and spatial remodeling that drives disease which, in turn, could provide new avenues for intervention and further study.

AI snapshot

Written from the paper’s full text.

TL;DR

This study combines MERFISH spatial transcriptomics and snRNA-seq to map healthy and fibrotic human liver at single-cell resolution, profiling ~310,000 cells and 317 genes. It defines continuous hepatocyte zonation, spatially distinct macrophage and HSC subsets, zonally enriched receptor-ligand interactions, and two injury-expanded hepatocyte populations without clear zonal distributions.

Key findings

  • Healthy liver MERFISH (~90,000 putative cells) resolved three hepatocyte clusters mapping to zones 1–3: SDS/CYP2A6 enriched in zone 1, ALDOB/ADH1B in zone 2, and CYP2E1/CYP1A2 in zone 3. Pseudotime and lobule-level gradients indicated hepatocyte zonation is better described as a continuous portal-to-central variable than as discrete zones.
  • Non-parenchymal cells included two macrophage clusters (Mac 1 periportal-enriched; Mac 2 diffuse, CD5L+/MARCO+, Kupffer-like) and two HSC clusters (HSC 1 periportal-enriched; HSC 2 diffuse; differing mainly in CD74). Cholangiocytes were KRT7+ portal cells and ECs were DNASE1L3+ diffuse; no clear non-parenchymal zonation was observed.
  • Approximately one third of hepatocytes were multinucleated; their distribution did not change across zones, and no significant gene-expression differences were found between hepatocytes with one versus two nuclei. Multinucleated hepatocytes were larger and had more total RNA counts, but RNA density did not depend on nuclei number.
  • In fibrosis, MERFISH profiled ~220,000 cells and snRNA-seq ~13,500 nuclei. Two hepatocyte populations (Fibrotic Hep 1 and 2) expanded with injury and lacked clear zonal distribution, while portal and central hepatocytes retained zonation; fibrotic macrophages lost MARCO, CD5L, and CD68, and fibrotic HSCs maintained high COL1A1. Urea-pathway genes including CPS1, SLC7A2, and SLC25A15 were enriched with injury.

Methods

  • Paired MERFISH and snRNA-seq were performed on the same healthy and fibrotic human liver samples. MERFISH used a 317-gene panel focused on hepatocytes plus non-parenchymal markers, with Na+/K+-ATPase immunofluorescence for cell-boundary staining.
  • Healthy tissue came from margins of surgical resections from three adult patients (two female, one male); fibrotic samples came from three additional donors. The study reports ~90,000 healthy MERFISH cells, ~15,000 healthy snRNA-seq nuclei, ~220,000 fibrotic MERFISH cells, and ~13,500 fibrotic snRNA-seq nuclei.
  • Cells were segmented with Cellpose and Baysor, analyzed in Scanpy with Harmony batch correction and Leiden clustering, and integrated across MERFISH and snRNA-seq. Receptor-ligand interactions were inferred with CellPhoneDB; differential expression, GO, and GSEA were used for injury comparisons.

Limitations

  • Stated in the paper: MERFISH covers only 317 genes, not the full transcriptome; the panel was not designed to explore diverse non-parenchymal pathways across zones, so the absence of observed non-parenchymal zonation may reflect panel content. Receptor-ligand results are hypothesis-generating only. Ploidy was not quantified separately, though the authors predict >80% of multinucleated hepatocytes are 2N. Small sample size precluded considering sex as a variable, and segmentation challenges were noted for HSC/macrophage markers in cholangiocytes. Lymphocyte and neutrophil populations were not defined because of insufficient marker genes in the panel.
  • Evident from the text: healthy samples were resection margins without clear signs of acute or chronic injury rather than confirmed disease-free donor liver, and the small donor numbers limit generalizability. The experiments were not randomized and investigators were not blinded.

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

Integrity signals

Registry

Crossref · 10.1038/s41467-024-55325-4

Cites

70 of 70 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

39 of 70 cited sources checked · 14 supported · 0 flagged · accuracy not yet measured

  • Findingcitation

    DNASE13 and INMT expression were scattered through the parenchyma, consistent with the distribution of ECs.

    “KRT7 and SOX9 were enriched in periportal areas consistent with cholangiocytes and include clusters suggestive of longitudinal sections of small bile ducts (Supplementary Fig. 2g–l), while DNASE13 and INMT expression were scattered through the parenchyma consistent with the distribution of ECs10 (Supplementary Fig. 2g–l).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [10] MacParland, S. A. et al. Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations. Nat. Commun. 9, 4383 (2018). source

      The Results section, “Liver endothelial cells” (page unavailable), lists INMT and DNASE1L3 among genes enriched in an endothelial-cell cluster, supporting the endothelial-cell association. It does not show that expression of either gene was scattered through the parenchyma; its spatial staining analysis concerns CD32B. The cited gene is written DNASE1L3, rather than DNASE13, in our copy. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
  • Premisecitation

    Sub-populations of macrophages and HSCs in the healthy human liver have been described in single-cell data.

    “While sub-populations of macrophages and HSCs in the healthy human liver have been described in single-cell data10–12, these subpopulations, and potential gene expression variations within, have not been resolved within the lobule organization at single-cell resolution in the human liver.”
    Quote in the paperSubstantial claim3 of 3 sources checked
    • cites [10] MacParland, S. A. et al. Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations. Nat. Commun. 9, 4383 (2018). source

      The Introduction reports two distinct liver-resident macrophage populations identified by single-cell RNA sequencing. But the Results section on hepatic stellate cells identifies only one HSC cluster (Cluster 20), not HSC subpopulations. The copy supplies section labels but no page numbers. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
    • cites [11] Ramachandran, P. et al. Resolving the fibrotic niche of human liver cirrhosis at single-cell level. Nature 38, 1–29 (2019). source

      The Results section on macrophages identifies distinct Kupffer-cell clusters in human liver samples that included healthy livers, supporting the macrophage portion. The Results section on mesenchymal cells identifies one cluster as hepatic stellate cells and distinguishes it from other mesenchymal populations, but does not describe multiple HSC subpopulations in healthy human liver. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [12] Andrews, T. S. et al. Single‐Cell, Single‐Nucleus, and Spatial RNA Sequencing of the Human Liver Identifies Cholangiocyte and Mesenchymal Heterogeneity. Hepatol. Commun. 6, 821–840 (2021). source
  • Findingcitation

    CD5L expression was clustered in spaces between hepatocytes, consistent with resident macrophages located in sinusoids.

    “We observed sheets of cells expressing ALDOB, consistent with hepatocytes21 (Supplementary Fig. 2), while CD5L expression was clustered in spaces between hepatocytes consistent with resident macrophages located in sinusoids10 (Supplementary Fig. 2a–f).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [10] MacParland, S. A. et al. Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations. Nat. Commun. 9, 4383 (2018). source
  • Findingown data

    ALDOB expression was observed in sheets of cells consistent with hepatocytes.

    “We observed sheets of cells expressing ALDOB, consistent with hepatocytes21 (Supplementary Fig. 2), while CD5L expression was clustered in spaces between hepatocytes consistent with resident macrophages located in sinusoids10 (Supplementary Fig. 2a–f).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [21] Aizarani, N. et al. A human liver cell atlas reveals heterogeneity and epithelial progenitors. Nature 472, 199–204 (2019). source

      The Zonation of human liver cell types section identifies other hepatocyte-associated genes, including ALB and PCK1, but does not mention ALDOB. Searches of our copy for ALDOB, aldolase B, and hepatocyte sheets found no passage linking ALDOB expression to hepatocytes or describing the sheet-like pattern. (LLM, conf=0.70)

      Experimental — accuracy not yet measured
  • Interpretationown data

    Spatial gene expression in human hepatocytes is better described by a continuous variable determined by the relative distance between the portal and central areas of the lobule, similar to observations in mouse.

    “Collectively, these observations suggest that spatial gene expression in human hepatocytes is better described by a continuous variable determined by the relative distance between the portal and central areas of the lobule, similar to observations in mouse7,26,27.”
    Quote in the paperSubstantial claim3 of 3 sources checked
    • cites [7] Droin, C. et al. Space-time logic of liver gene expression at sub-lobular scale. Nat. Metab. 3, 43–58 (2021). source

      The Results section describes spatial gene-expression profiles along the central–portal axis in mouse hepatocytes, supporting the comparison to mouse zonation. The study assigns cells to eight lobule layers using gene-expression landmarks; it does not establish that human hepatocyte expression is better described by a continuous variable determined by relative distance from the portal and central areas. Page numbers are unavailable in our copy. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites [26] Halpern, K. B. et al. Single-cell spatial reconstruction reveals global division of labour in the mammalian liver. Nature 542, 352–356 (2017). source
    • cites [27] González-Blas, C. B. et al. Single-cell spatial multi-omics and deep learning dissect enhancer-driven gene regulatory networks in liver zonation. Nat. Cell Biol. 26, 153–167 (2018). source

      The 'Zonation drives cell state heterogeneity in hepatocytes' section reports a gene-expression gradient along the portocentral axis, and Fig. 1 identifies the atlas as mouse liver (page not provided). This supports the mouse comparison, but the cited paper does not establish that the same continuous description applies to human hepatocytes. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
  • Classified as probably not a substantial claim. 30
  • May combine two claims. 73
  • The quote only partly states this claim. 2

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

    Premisecitation

    The liver lobules are 0.5-1 mm in diameter and defined by sites of portal venous and arterial inflow and central venous outflow.

    “The liver is composed of parenchymal, non-parenchymal, and immune cells that are organized into anatomic structures called lobules, which are 0.5–1 mm in diameter and defined by sites of portal venous and arterial inflow and central venous outflow1,2.”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [1] Rezania, V., Coombe, D. & Tuszynski, J. A. A physiologically-based flow network model for hepatic drug elimination III: 2D/3D DLA lobule models. Theor. Biol. Med. Model 13, 9 (2016). source

      The Background section, “The liver lobule functional unit” (page unavailable), describes a classic lobule with a central hepatic venule and portal tracts at its corners. The Background also mentions flow from portal vein and hepatic artery segments to a hepatic vein in a model it discusses. Our copy does not give a 0.5–1 mm lobule diameter; its Methods describe a modeled diameter of 1.536 mm. Thus the vascular description is supported, but the stated diameter is not. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
    • cites [2] Teutsch, H. F. The modular microarchitecture of human liver. Hepatology 42, 317–325 (2005). source

  • May combine two claims.

    Findingcitation

    SDS was enriched in zone 1 (periportal), while CYP2E1 expression increased in zone 3 (pericentral), as previously described in mouse.

    “SDS was enriched in zone 1 (periportal), while CYP2E1 expression increased in zone 3 (pericentral), as previously described in mouse24–27 (Fig. 1c, e, f; Supplementary Fig. 5a).”
    Quote in the paperSubstantial claim3 of 4 sources checked
    • cites [24] Braeuning, A. et al. Differential gene expression in periportal and perivenous mouse hepatocytes. Febs J. 273, 5051–5061 (2006). source
    • cites [25] Saito, K., Negishi, M. & Squires, E. J. Sexual dimorphisms in zonal gene expression in mouse liver. Biochem. Biophys. Res. Commun. 436, 730–735 (2013). source
    • cites [26] Halpern, K. B. et al. Single-cell spatial reconstruction reveals global division of labour in the mammalian liver. Nature 542, 352–356 (2017). source

      The Extended Data Figure 9 caption reports a pericentral bias for liver detoxification genes, including cytochrome P450 genes, but does not identify CYP2E1. Searches for SDS and CYP2E1 found no passages in our copy stating either gene’s zonation, so the specific mouse findings attributed to this paper cannot be verified. (LLM, conf=0.72)

      Experimental — accuracy not yet measured
    • cites [27] González-Blas, C. B. et al. Single-cell spatial multi-omics and deep learning dissect enhancer-driven gene regulatory networks in liver zonation. Nat. Cell Biol. 26, 153–167 (2018). source

      In “Zonation drives cell state heterogeneity in hepatocytes” (page unavailable), the paper identifies a Cyp2e1-positive zone encompassing pericentral and mid-lobular mouse hepatocytes, supporting the pericentral part of the claim. The Fig. 1 caption names Sds among genes imaged by smFISH but does not state that it is enriched in periportal zone 1. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

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

    Premisecitation

    Morphogen expression, such as that of Wnt-family members, is shaped by hepatocyte zonation and can vary in time.

    “Beyond metabolism, morphogen expression, such as that of Wnt-family members, is also shaped by this zonation and can vary in time7.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [7] Droin, C. et al. Space-time logic of liver gene expression at sub-lobular scale. Nat. Metab. 3, 43–58 (2021). source

      The Results section reports that Wnt2 expression in pericentral non-parenchymal liver cells varies around the clock, supporting the temporal part of the claim. The Discussion describes Wnt ligands as forming a spatial gradient that contributes to zonated Wnt-target expression in hepatocytes; it does not show that hepatocyte zonation shapes morphogen expression. (LLM, conf=0.85)

      Experimental — accuracy not yet measured

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

    Premisecitation

    Hepatocytes in zone 3 are more active in glycolysis and lipogenesis, reflecting the depletion of oxygen and nutrients near the central vein.

    “For example, hepatocytes in zone 1 are more involved in gluconeogenesis and beta-oxidation, reflecting the relatively oxygen- and nutrient-rich environment near the portal vein and hepatic artery, whereas hepatocytes in zone 3 are more active in glycolysis and lipogenesis, reflecting the depletion of oxygen and nutrients near the central vein3,4,6.”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites [3] Gebhardt, R. Metabolic zonation of the liver: Regulation and implications for liver function. Pharm. Therapeut 53, 275–354 (1992). source
    • cites [4] Jungermann, K. & Keitzmann, T. Zonation of Parenchymal and Nonparenchymal Metabolism in Liver. Annu Rev. Nutr. 16, 179–203 (1996). source
    • cites [6] Kietzmann, T. Metabolic zonation of the liver: The oxygen gradient revisited. Redox Biol. 11, 622–630 (2017). source

      The 'Factors involved in the regulation of zonation' section says oxygen falls toward the perivenous region and that blood flow creates substrate gradients. 'Hedgehog signaling' reports that FASN, a lipogenic enzyme, is perivenous in control mice, supporting the lipogenesis portion. The copy does not establish that zone 3 hepatocytes are more active in glycolysis or that depletion of both oxygen and nutrients explains both activities. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

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

    Premisecitation

    Hepatocytes in zone 1 are more involved in gluconeogenesis and beta-oxidation, reflecting the relatively oxygen- and nutrient-rich environment near the portal vein and hepatic artery.

    “For example, hepatocytes in zone 1 are more involved in gluconeogenesis and beta-oxidation, reflecting the relatively oxygen- and nutrient-rich environment near the portal vein and hepatic artery, whereas hepatocytes in zone 3 are more active in glycolysis and lipogenesis, reflecting the depletion of oxygen and nutrients near the central vein3,4,6.”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites [3] Gebhardt, R. Metabolic zonation of the liver: Regulation and implications for liver function. Pharm. Therapeut 53, 275–354 (1992). source
    • cites [4] Jungermann, K. & Keitzmann, T. Zonation of Parenchymal and Nonparenchymal Metabolism in Liver. Annu Rev. Nutr. 16, 179–203 (1996). source
    • cites [6] Kietzmann, T. Metabolic zonation of the liver: The oxygen gradient revisited. Redox Biol. 11, 622–630 (2017). source

      The section “Hypoxia, HIFs, and hedgehog signaling: another liaison” says fatty-acid oxidation is more pronounced in periportal hepatocytes under starvation. “Factors involved in the regulation of zonation” describes higher oxygen levels near the portal vein and hepatic artery and gradients of substrates along the blood flow. Our copy does not specifically place gluconeogenesis in zone 1, and the oxidation finding carries a starvation qualifier absent from the claim. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Study design and conductown data

    These observations are consistent with the classic zonation described in human and mouse liver, supporting the MERFISH measurements.

    “Collectively, these observations are consistent with the classic zonation described in human and mouse liver4,24–26, supporting our MERFISH measurements.”
    Quote in the paperSubstantial claim2 of 4 sources checked
    • cites [4] Jungermann, K. & Keitzmann, T. Zonation of Parenchymal and Nonparenchymal Metabolism in Liver. Annu Rev. Nutr. 16, 179–203 (1996). source
    • cites [24] Braeuning, A. et al. Differential gene expression in periportal and perivenous mouse hepatocytes. Febs J. 273, 5051–5061 (2006). source
    • cites [25] Saito, K., Negishi, M. & Squires, E. J. Sexual dimorphisms in zonal gene expression in mouse liver. Biochem. Biophys. Res. Commun. 436, 730–735 (2013). source
    • cites [26] Halpern, K. B. et al. Single-cell spatial reconstruction reveals global division of labour in the mammalian liver. Nature 542, 352–356 (2017). source

      The main text describes established liver zonation and reports zonation profiles in mouse liver; it also says those profiles correlated with earlier datasets. This supports the mouse-liver portion of the cited background, but the paper does not establish zonation in human liver. The statement that this supports the citing authors’ MERFISH measurements is their own inference. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingcitation

    PDGFRB expression was concentrated along the edges of ALDOB-expressing hepatocytes in a distribution expected for HSCs, which are located in the subendothelial space.

    “PDGFRB expression was concentrated along the edges of ALDOB-expressing hepatocytes in a distribution expected for HSCs, which are located in the subendothelial space11 (Supplementary Fig. 2a–f).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [11] Ramachandran, P. et al. Resolving the fibrotic niche of human liver cirrhosis at single-cell level. Nature 38, 1–29 (2019). source

      The citation is attached to the statement that hepatic stellate cells occupy the subendothelial space. The Results section on mesenchymal cells identifies an HSC population by RGS5 staining, but our copy does not state its subendothelial location. Searches for “subendothelial” and “space of Disse” found no passage establishing that location; the PDGFRB–ALDOB staining is attributed to the citing paper’s supplementary figure. (LLM, conf=0.70)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingown data

    No clear patterns of zonation in gene expression within non-parenchymal populations were noted, suggesting that if such gradients exist in the human liver, the MERFISH library does not contain the proper genes to capture them.

    “We did not note clear patterns of zonation in gene expression within these non-parenchymal populations28–30, suggesting that, if such gradients exist in the human liver, our MERFISH library does not contain the proper genes to capture them.”
    Quote in the paperNot a substantial claim3 of 3 sources checked
    • cites [28] Halpern, K. B. et al. Paired-cell sequencing enables spatial gene expression mapping of liver endothelial cells. Nat. Biotechnol. 36, 962–970 (2018). source
    • cites [29] Dobie, R. et al. Single-Cell Transcriptomics Uncovers Zonation of Function in the Mesenchyme during Liver Fibrosis. Cell Rep. 29, 1832–1847.e8 (2019). source
    • cites [30] Inverso, D. et al. A spatial vascular transcriptomic, proteomic, and phosphoproteomic atlas unveils an angiocrine Tie–Wnt signaling axis in the liver. Dev. Cell 56, 1677–1693.e10 (2021). source

  • May combine two claims.

    Findingcitation

    Both macrophage populations were identified by expression of CD74, while Mac 2 cells also expressed CD5L and MARCO, most consistent with non-inflammatory macrophages or Kupffer cells.

    “Both macrophage populations were identified by expression of CD74, while Mac 2 cells also expressed CD5L and MARCO (Fig. 2d, i; Supplementary Fig. 6b), most consistent with non-inflammatory macrophages or Kupffer cells10,11,21.”
    Quote in the paperNot a substantial claim3 of 3 sources checked
    • cites [10] MacParland, S. A. et al. Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations. Nat. Commun. 9, 4383 (2018). source
    • cites [11] Ramachandran, P. et al. Resolving the fibrotic niche of human liver cirrhosis at single-cell level. Nature 38, 1–29 (2019). source
    • cites [21] Aizarani, N. et al. A human liver cell atlas reveals heterogeneity and epithelial progenitors. Nature 472, 199–204 (2019). source

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

    Premisecitation

    CD74 is expressed by HSCs, can be induced by inflammatory signals such as IFN-γ, and can promote immune activation through MHC class II-mediated interactions.

    “CD74 is expressed by HSCs, can be induced by inflammatory signals such as IFN-γ and can promote immune activation through MHC class II-mediated interactions46,47.”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [46] Maubach, G., Lim, M. C. C., Kumar, S. & Zhuo, L. Expression and upregulation of cathepsin S and other early molecules required for antigen presentation in activated hepatic stellate cells upon IFN-γ treatment. Biochim. Biophys. Acta (BBA) - Mol. Cell Res. 1773, 219–231 (2007). source
    • cites [47] Viñas, O. et al. Human hepatic stellate cells show features of antigen‐presenting cells and stimulate lymphocyte proliferation. Hepatology 38, 919–929 (2003). source

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

    Premisecitation

    THRB encodes the protein target of resmetirom, a medication shown to be effective in treating nonalcoholic steatohepatitis.

    “THRB, which encodes the protein target of resmetirom, a medication shown to be effective in treating nonalcoholic steatohepatitis42”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [42] Harrison, S. A. et al. A Phase 3, Randomized, Controlled Trial of Resmetirom in NASH with Liver Fibrosis. N. Engl. J. Med. 390, 497–509 (2024). source

      The abstract (p. 2) identifies thyroid hormone receptor beta as resmetirom’s target, and the Efficacy section (p. 9) reports better NASH-resolution and fibrosis outcomes than placebo at week 52. Our copy does not state that the THRB gene encodes that receptor, so the gene-to-target part of the claim is not directly supported. (LLM, conf=0.82)

      Experimental — accuracy not yet measured

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

    Interpretationcitation

    The THBS1/2-CD36 interaction may influence MASLD.

    “which may influence metabolic dysfunction associated steatotic liver disease (MASLD)48”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [48] Zeng, H. et al. CD36 promotes de novo lipogenesis in hepatocytes through INSIG2-dependent SREBP1 processing. Mol. Metab. 57, 101428 (2022). source

      The Introduction and Discussion link CD36 to nonalcoholic fatty liver disease, and the Results report that hepatocyte-specific CD36 deletion protected mice from high-fat-diet-induced steatosis. But the interaction examined in the Results is CD36–INSIG2; our copy does not establish a THBS1/2–CD36 interaction or its effect on liver disease (searched: THBS1, THBS2, thrombospondin). (LLM, conf=0.87)

      Experimental — accuracy not yet measured

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

    Premisecitation

    PNPLA3, CPEB4, and ASPG are linked to hepatic steatosis, steatohepatitis, and/or fibrosis.

    “PNPLA3, CPEB4, and ASPG, which are linked to hepatic steatosis, steatohepatitis, and/or fibrosis33,39–41”
    Quote in the paperNot a substantial claim3 of 4 sources checked
    • cites [33] Romeo, S. et al. Genetic variation in PNPLA3 confers susceptibility to nonalcoholic fatty liver disease. Nat. Genet. 40, 1461–1465 (2008). source

      The paper reports that a PNPLA3 variant is strongly associated with hepatic fat content and that carriers have elevated markers of liver inflammation (Abstract; page not provided). Our copy does not link CPEB4 or ASPG to these conditions, and it does not establish a PNPLA3 association with fibrosis. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
    • cites [39] Valenti, L. et al. Homozygosity for the patatin‐like phospholipase‐3/adiponutrin I148M polymorphism influences liver fibrosis in patients with nonalcoholic fatty liver disease. Hepatology 51, 1209–1217 (2010). source
    • cites [40] Maillo, C. et al. Circadian- and UPR-dependent control of CPEB4 mediates a translational response to counteract hepatic steatosis under ER stress. Nat. Cell Biol. 19, 94–105 (2017). source
    • cites [41] Xanthakos, S. A. et al. High Prevalence of Nonalcoholic Fatty Liver Disease in Adolescents Undergoing Bariatric Surgery. Gastroenterology 149, 623–634.e8 (2015). source

      The Results (Hepatic Gene Expression Profiles; page unavailable) report gene-expression differences across NAFLD and NASH groups but do not identify PNPLA3, CPEB4, or ASPG. Searches for those three gene names found no passage in our copy linking them to steatosis, steatohepatitis, or fibrosis. (LLM, conf=0.72)

      Experimental — accuracy not yet measured

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

    Premisecitation

    SLC25A15 is an ornithine transporter in the urea cycle.

    “SLC25A15, is an ornithine transporter in the urea cycle38”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [38] Fiermonte, G. et al. The Mitochondrial Ornithine Transporter Bacterial Expression, Reconstitution, Functional Characterization, And Tissue Distribution Of Two Human Isoforms*. J. Biol. Chem. 278, 32778–32783 (2003). source

  • May combine two claims.

    Findingown data

    CYP2A6 and ASS1 were enriched in portal regions similar to SDS, while CYP1A2 and GLUL were enriched in central regions similar to CYP2E1.

    “In addition, we found that CYP2A6 and ASS1 were enriched in portal regions similar to SDS, while CYP1A2 and GLUL were enriched in central regions similar to CYP2E1 (Fig. 1c, e, f–g).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    MERFISH and snRNA-seq analysis of fibrotic liver samples identify two hepatocyte populations that expand with injury and do not have clear zonal distributions.

    “Finally, MERFISH and snRNA-seq analysis of fibrotic liver samples identify two hepatocyte populations that expand with injury and do not have clear zonal distributions.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Mapping hepatocyte clusters 1, 2, and 3 back to tissue sections showed clear zonal distribution, with hepatocyte cluster 1 localizing to zone 1, hepatocyte cluster 3 to zone 3, and hepatocyte cluster 2 to zone 2.

    “Mapping hepatocyte clusters 1, 2, and 3 (Fig. 1a) back to tissue sections showed clear zonal distribution, with hepatocyte cluster 1 localizing to zone 1, hepatocyte cluster 3 to zone 3, and hepatocyte cluster 2 to zone 2 (between zone 1 and 3; Fig. 1h, Supplementary Fig. 5b, c).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingreasoning

    Hepatocyte clusters contained more RNA counts and more expressed genes than other clusters.

    “Reflecting both the hepatocyte focus of our MERFISH panel as well as the larger size of hepatocytes, hepatocyte clusters contained more RNA counts and more expressed genes than other clusters (Supplementary Fig. 4b–d).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    KRT7 and SOX9 were enriched in periportal areas consistent with cholangiocytes and included clusters suggestive of longitudinal sections of small bile ducts.

    “KRT7 and SOX9 were enriched in periportal areas consistent with cholangiocytes and include clusters suggestive of longitudinal sections of small bile ducts (Supplementary Fig. 2g–l), while DNASE13 and INMT expression were scattered through the parenchyma consistent with the distribution of ECs10 (Supplementary Fig. 2g–l).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Three clusters of hepatocytes were defined along with two HSC clusters, two macrophage clusters, one EC cluster, and one cholangiocyte cluster.

    “Three clusters of hepatocytes were defined along with two HSC clusters, two macrophage clusters, one EC cluster, and one cholangiocyte cluster.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    ALDOB and ADH1B were expressed in all hepatocytes but were enriched in zone 2, with ALDOB expression shifted towards zones 1 and 2 and ADH1B expression shifted towards zones 2 and 3.

    “ALDOB and ADH1B were expressed in all hepatocytes but were enriched in zone 2, with ALDOB expression shifted towards zones 1 and 2 and ADH1B expression shifted towards zones 2 and 3 (Fig. 1d, f).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Findingown data

    snRNA-seq analysis for the same human liver samples profiled approximately 15,000 nuclei after filtering.

    “To address this limitation, we next performed snRNA-seq analysis for the same human liver samples analyzed by MERFISH, profiling ~15,000 nuclei after filtering (Supplementary Fig. 7a).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Macrophage cluster 1 was enriched in the periportal area and also dispersed through the lobules, while cells from Mac 2 were scattered more diffusely through the lobules.

    “Macrophage (Mac) cluster 1 was enriched in the periportal area and also dispersed through the lobules, while cells from Mac 2 were scattered more diffusely through the lobules.”
    Quote in the paperNot a substantial claim

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

    Findingown data

    In total, approximately 220,000 cells were imaged from fibrotic liver samples.

    “We leveraged the same protocols as described above to image and segment cells and, in total, imaged ~220,000 cells (Supplementary Fig. 9).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Joint clustering of hepatocytes in the snRNA-seq data with those measured in MERFISH identified hepatocyte zones 1, 2, and 3 as seen in the MERFISH data, but the axis of zonation within hepatocytes was not as clear in the snRNA-seq data as it was in the MERFISH measurements.

    “Visualization of the snRNA-seq data after this integration showed that we could identify hepatocytes zones 1, 2, and 3 as seen in the MERFISH data, but the axis of zonation within hepatocytes was not as clear in the snRNA-seq data as it was in the MERFISH measurements (Fig. 1a; Fig. 3a).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Findingown data

    snRNA-seq on the same samples characterized with MERFISH collected approximately 13,500 nuclei across three samples.

    “To expand our analysis transcriptome-wide, we next performed snRNA-seq on the same samples characterized with MERFISH, collecting ~13,500 nuclei across these three samples (Fig. 6; Supplementary Fig. 11).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    HSC 1 and ECs expressed EFNA5 and EFNB2, respectively, to signal to portal hepatocytes through EPHA2 and EPHB1.

    “TF produced by Mac 1 cells can signal to TFRC on periportal hepatocytes, while HSC 1 and ECs expressed EFNA5 and EFNB2, respectively to signal to portal hepatocytes through EPHA2 and EPHB1.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Co-expression of receptor-ligand pairs in snRNA-seq data indicates potential cell-cell interactions in situ.

  • May combine two claims.

    Findingown data

    Leveraging the snRNA-seq alone defined two populations of HSCs, two populations of macrophages, a population of cholangiocytes, and a population of ECs.

    “Leveraging the snRNA-seq alone, we also defined two populations of HSCs, two populations of macrophages, a population of cholangiocytes, and a population of ECs (Fig. 3a, b; Supplementary Data 2).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    VEGFA, expressed by pericentral hepatocytes, is recognized by NRP2, expressed by HSC 1 cells, and NRP1, expressed by Mac 2 cells.

    “VEGFA, expressed by pericentral hepatocytes is recognized by NRP2, expressed by HSC 1 cells and NRP1 expressed by Mac 2 cells.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Co-expression of receptor-ligand pairs in snRNA-seq data indicates potential cell-cell interactions in situ.

  • May combine two claims.

    Findingown data

    Portal and central hepatocytes from the fibrotic liver retain the same zonal expression of marker genes as healthy zone 1, 2, and 3 and are characterized primarily by a reduction in expression of these genes.

    “Portal and central hepatocytes from the fibrotic liver retain the same zonal expression of marker genes as healthy zone 1, 2, and 3 and are characterized primarily by a reduction in expression of these genes.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    SEMA4D expressed by Mac 1 and Mac 2 cells, and VEGFB produced by Mac 2 cells, can be recognized by PLXNB2 and NRP1, respectively, both of which are expressed by pericentral hepatocytes.

    “GAS6 produced by HSC 1 and ECs can signal to pericentral hepatocytes through MERTK, while SEMA4D expressed by Mac 1 and Mac 2 cells, and VEGFB produced by Mac 2 cells can be recognized by PLXNB2 and NRP1, respectively, both of which are expressed by pericentral hepatocytes.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Co-expression of receptor-ligand pairs in snRNA-seq data indicates potential cell-cell interactions in situ.

  • May combine two claims.

    Findingown data

    Fibrotic macrophages showed a loss of MARCO, CD5L, CD68 and retained an expression pattern that more closely matched Mac 1 cells in normal liver.

    “For example, fibrotic macrophages showed a loss of MARCO, CD5L, CD68 and retained an expression pattern that more closely matched Mac 1 cells in normal liver (Fig. 5g), while fibrotic HSCs maintained high levels of COL1A1 expression without a clear positive signature from the other genes included in the MERFISH panel (Fig. 5h).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Joint analysis of MERFISH data across all samples produced two hepatocyte clusters: a periportal cluster of hepatocytes (Portal Hep) and a pericentral cluster of hepatocytes (Central Hep), which collectively contained zone 1, 2 and 3 hepatocytes from healthy liver.

    “Instead of three zones of hepatocytes, this joint analysis produced two: a periportal cluster of hepatocytes (Portal Hep) and a pericentral cluster of hepatocytes (Central Hep), which collectively contained zone 1, 2 and 3 hepatocytes from healthy liver (Fig. 5a, Supplementary Fig. 9b-d, Supplementary Fig. 10, Supplementary Data 2).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Cholangiocytes and ECs were identified by MERFISH and followed predicted distributions.

    “Cholangiocytes and ECs were also identified by MERFISH and followed predicted distributions (Fig. 2a, b, g; Supplementary Fig. 6e).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Some evidence of lower HSC and macrophage marker expression in the cholangiocytes seen with snRNA-seq relative to those seen with MERFISH was noted, consistent with modest segmentation challenges with these cells.

    “However, we did note some evidence of lower HSC and macrophage marker expression in the cholangiocytes seen with snRNA-seq relative to those seen with MERFISH, consistent with modest segmentation challenges with these cells (Supplementary Fig. 8d).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    HSC 1 cells were enriched in periportal regions and spread through the lobules, while HSC 2 cells were scattered more diffusely through the lobules without periportal enrichment.

    “These cells also displayed different spatial distributions, with cells from HSC 1 enriched in periportal regions and spread through the lobules, while cells from HSC 2 were scattered more diffusely through the lobules without periportal enrichment.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    ECs and Mac 2 cells also produce WNT2B, which can be recognized by FZD6-LRP5/6 in periportal hepatocytes, with the highest interaction scores observed between ECs and zone 3 hepatocytes.

    “ECs and Mac 2 cells also produce WNT2B, which can be recognized by FZD6-LRP5/6 in periportal hepatocytes, with the highest interaction scores observed between ECs and zone 3 hepatocytes (Supplementary Data 3).”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Co-expression of receptor-ligand pairs in snRNA-seq data indicates potential cell-cell interactions in situ.

  • May combine two claims.

    Findingown data

    HSCs displayed bands of cells across lobules, while macrophages were spread more diffusely through the lobules.

    “HSCs displayed bands of cells across lobules, while macrophages were spread more diffusely through the lobules (Fig. 5e).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Two macrophage populations were visualized in the healthy liver and showed different patterns of distribution.

    “Two macrophage populations were visualized in the healthy liver and showed different patterns of distribution (Fig. 2a, b, c, Supplementary Fig. 6a).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    ECs, many of which line the sinusoids, were identified more diffusely through the liver, and expressed DNASE1L3.

    “Cholangiocytes were localized to portal regions and were identified by expression of KRT7 (Fig. 2h, left; Supplementary Fig. 6f, left), while ECs, many of which line the sinusoids, were identified more diffusely through the liver, and expressed DNASE1L3 (Fig. 2h, right; Supplementary Fig. 6f, right).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Cholangiocytes were localized to portal regions and were identified by expression of KRT7.

    “Cholangiocytes were localized to portal regions and were identified by expression of KRT7 (Fig. 2h, left; Supplementary Fig. 6f, left), while ECs, many of which line the sinusoids, were identified more diffusely through the liver, and expressed DNASE1L3 (Fig. 2h, right; Supplementary Fig. 6f, right).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    A single cluster of HSCs and a single cluster of macrophages were also identified.

    “A single cluster of HSCs and a single cluster of macrophages were also identified (Fig. 5a, b; Supplementary Data 2).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Hepatocytes can be separated into three populations, but these populations are contained within a single group of cells in the UMAP representation of the data.

    “While our analysis revealed that hepatocytes can be separated into three populations, we noted that these populations are contained within a single group of cells in the UMAP representation of the data (Fig. 1a).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Joint integration and clustering of hepatocytes between MERFISH and snRNA-seq revealed nuclei in the snRNA-seq data that were similar to the two fibrotic hepatocyte clusters (Fibrotic Hep 1 and Fibrotic Hep 2) seen in the MERFISH measurements, supporting the emergence of these two disease-associated populations.

    “Importantly, this analysis revealed nuclei in the snRNA-seq data that were similar to the two fibrotic hepatocyte clusters (Fibrotic Hep 1 and Fibrotic Hep 2) seen in the MERFISH measurements, supporting the emergence of these two disease-associated populations (Supplementary Fig. 12a, b).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The two HSC populations differed primarily in the levels of CD74 expression, while both populations expressed PDGFRB and LAMB1.

    “These two populations differed primarily in the levels of CD74 expression, while both populations expressed PDGFRB and LAMB1 (Fig. 2f, i; Supplementary Fig. 6d).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Total RNA counts per cell increased with nuclei number, as did cell area, such that multinucleated hepatocytes tended to be larger and contain proportionally more RNA transcripts, but RNA density was not dependent on the number of nuclei.

    “Nonetheless, the total RNA counts per cell increased with nuclei number (Fig. 4f, left) as did the cell area (Fig. 4f, center), such that multinucleated hepatocytes tended to be larger and contain proportionally more RNA transcripts, but RNA density was not dependent on the number of nuclei (Fig. 4f, right).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    THBS1/2 is detected at the highest level in HSCs and can signal to CD36 expressed by pericentral hepatocytes, while the THBS1-CD36 receptor-ligand pair is also enriched between pericentral hepatocytes and macrophages.

    “THBS1/2 is detected at the highest level in HSCs and can signal to CD36 expressed by pericentral hepatocytes (Supplementary Data 3), while the THBS1-CD36 receptor-ligand pair is also enriched between pericentral hepatocytes and macrophages.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Co-expression of receptor-ligand pairs in snRNA-seq data indicates potential cell-cell interactions in situ.

  • May combine two claims.

    Findingown data

    Combining MERFISH with snRNA-seq extended transcriptional profiles to understand unique receptor-ligand interactions involving hepatocytes and non-parenchymal cells across hepatocyte zones and defined changes in gene expression in spatially-resolved hepatocyte populations with injury.

    “By combining MERFISH with snRNA-seq, we extended the transcriptional profiles of cell types in these spatial maps to understand unique receptor-ligand interactions involving hepatocytes and non-parenchymal cells across hepatocyte zones and defined changes in gene expression in spatially-resolved hepatocyte populations with injury.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    With fibrotic injury, the macrophage population loses expression of MARCO, CD5L, and CD68 in MERFISH data, and snRNA-seq shows reduction in MARCO, CD5L, VCAM1, and CD163.

    “With the development of fibrotic injury, the macrophage population lost expression of MARCO, CD5L, and CD68 in the MERFISH data, and snRNA-seq analysis also showed reduction in MARCO and CD5L as well as VCAM1 and CD163.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Both Fibrotic Hep 1 and Fibrotic Hep 2 continue to show enriched expression of the nuclear receptors THRB, RORA, and ESR1 and maintain increased expression of CPS1 and SLC25A15.

    “Both Fibrotic Hep 1 and Fibrotic Hep 2 continue to show enriched expression of the nuclear receptors THRB, RORA, and ESR1 (Fig. 6c, Supplementary Data 4 and Supplementary Data 6) and maintain increased expression of CPS1 and SLC25A15.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Mac 2 cells are MARCO+, CD5L+, CD68+, with increased expression of VSIG4 compared to Mac 1 cells, and are located diffusely through the lobule.

    “Mac 2 cells were identified as MARCO+, CD5L+, and CD68+ with increased expression of VISG4 compared to Mac 1 cells and were located more diffusely through the lobule”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Two populations of macrophages (Mac 1 and Mac 2) were identified by MERFISH and show different distribution patterns through the lobule.

    “Two populations of macrophages were identified by MERFISH and show different distribution patterns through the lobule.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The transmembrane receptor EGFR is increased in Fibrotic Hep 1 and Fibrotic Hep 2 cells, while the transmembrane receptor GHR is increased in all injured hepatocyte populations, and LEPR is enriched in injured Portal Heps and Fibrotic Hep 1 and Fibrotic Hep 2 populations.

    “The transmembrane receptor EGFR is increased in Fibrotic Hep 1 and Fibrotic Hep 2 cells, while the transmembrane receptor GHR is increased in all injured hepatocyte populations, and LEPR is enriched in injured Portal Heps and Fibrotic Hep 1 and Fibrotic Hep 2 populations (Fig. 6c, Supplementary Data 4 and Supplementary Data 6).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Most transcripts identified in hepatocytes are cytoplasmic, but transcripts are also present in the nucleus.

    “Most of the transcripts identified in hepatocytes are cytoplasmic, but transcripts are also present in the nucleus.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    CD74 shows the greatest differential expression between the two HSC populations, with highest expression in HSC 1 cells.

    “the MHC class II chaperone CD7445 showed the greatest differential expression between the two HSC populations, with highest expression observed in HSC 1 cells.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Multinucleated hepatocytes tend to be larger in size, and mRNA levels are proportional to cell size.

    “In addition, multinucleated hepatocytes also tend to be larger in size, such that mRNA levels were proportional to cell size.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The study applied MERFISH to create spatial transcriptomic maps of healthy and fibrotic liver at single-cell resolution, defining spatially and transcriptionally distinct sub-populations of hepatocytes, macrophages, and HSCs within hepatic lobules and distinct hepatocyte sub-populations that expand with fibrotic injury.

    “Taken together, our study applied MERFISH to create spatial transcriptomic maps of the healthy and fibrotic liver at single-cell resolution to define spatially and transcriptionally distinct sub-populations of hepatocytes, macrophages, and HSCs within hepatic lobules and distinct hepatocyte sub-populations that expand with fibrotic injury.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The ligase CPS1 is increased in both populations, and GLUL remains more restricted to Central Heps and ASS1 to Portal Heps with injury.

    “The ligase CPS1 is increased in both populations, and GLUL remains more restricted to Central Heps and ASS1 to Portal Heps with injury.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Signaling interactions were identified including THBS1/2 from HSCs to CD36 in zone 3 hepatocytes, and WNT2B from ECs and Mac 2 cells with FZD6 and LRP5/6 in zone 3 hepatocytes.

    “we identified signaling interactions including a thrombospondin signal (THBS1/2) from HSCs to CD36 in zone 3 hepatocytes, which may influence metabolic dysfunction associated steatotic liver disease (MASLD)48, and the interaction of WNT2B, expressed by ECs and Mac 2 cells, with FZD6 and LPR5/6 expressed by zone 3 hepatocytes49,50 helps validate this analysis.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    HSC 1 is enriched in portal areas compared to HSC 2, and both populations are distributed through the lobules.

    “HSC 1 was enriched in the portal areas compared to HSC 2, and both populations were found distributed through the lobules.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Multinucleated hepatocytes do not show a change in relative gene expression with increase in nuclear content, but show an increase in RNA transcripts.

    “While we did not observe a change in relative gene expression with increase in nuclear content, we did observe an increase in RNA transcripts.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Multinucleated hepatocytes do not show differences in relative gene expression compared to mono-nucleated hepatocytes, but tend to be larger and produce more total transcripts.

    “Finally, by evaluating nuclear content, we found that multinucleated hepatocytes do not show differences in relative gene expression compared to mono-nucleated hepatocytes, but tend to be larger and produce more total transcripts.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Within the hepatic populations that emerge with fibrotic injury and do not show zonal distribution (Fibrotic Hep 1 and 2), an enrichment in metabolic pathways primarily involving fatty acids and amino acids was observed, while pathways involving RNA processing and ribosome biosynthesis are reduced.

    “Within the hepatic populations that emerge with fibrotic injury and do not show zonal distribution (Fibrotic Hep 1 and 2, Fig. 5d), we again observe an enrichment in metabolic pathways, primarily those involving fatty acids and amino acids, while pathways involving RNA processing and ribosome biosynthesis are reduced (Supplementary Fig 13a, b).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Fibrotic Hep 2 cells show relative enrichment in expression of hydrolases, heat shock proteins (HSPH1 and HSP90AA1, while HSPH1 is enriched in both Fibrotic Hep 1 and 2), and PLCB1 compared to Fibrotic Hep 1 cells.

    “Fibrotic Hep 2 cells show relative enrichment in expression of hydrolases, heat shock proteins (HSPH1 and HSP90AA1, while HSPH1 is enriched in both Fibrotic Hep 1 and 2), and PLCB1 compared to Fibrotic Hep 1 cells.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    snRNA-seq comparisons confirmed expression patterns and showed increased CD163, VCAM1, and TMEM26 in Mac 2 cells.

    “Further comparisons of gene expression between Mac1 and Mac 2 with snRNA-seq confirmed these expression patterns along with increased CD163, VCAM1, and TMEM26 in Mac 2 cells (Supplementary Data 2)”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Mac 1 cells are MARCO- and CD5L- with lower levels of CD68, and are enriched in portal areas in contrast to Mac 2 cells.

    “Mac 1 cells were MARCO- and CD5L- with lower levels of CD68. While Mac 1 cells were also found through the lobule, they did exhibit a distinct enrichment in portal areas in contrast to Mac 2 cells.”
    Quote in the paperSubstantial claim

  • The quote only partly states this claim.

    Findingown data

    A similar effect (CD74 differential expression) was observed in snRNA-seq data.

    “A similar effect was also observed in the snRNA-seq data.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In fibrotic injury, IL18 expression is maintained and gene expression remains distinct from the Mac 1 population.

    “IL18 expression was maintained, and gene expression remained distinct from the Mac 1 population (Fig. 6d).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    While oxygenases increase in both Portal and Central Heps with injury, CYP2E1 shows the highest expression in Central Heps with injury.

    “While oxygenases increase in both Portal and Central Heps with injury, CYP2E1 shows the highest expression in Central Heps with injury (Fig. 6c, Supplementary Data 4 and Supplementary Data 6).”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    Hepatocytes in the adult human liver can be single or multinucleated, and nuclei themselves can vary in ploidy.

    “Specifically, hepatocytes in the adult liver can either be single or multinucleated, and nuclei themselves can vary in ploidy.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [14] Kudryavtsev, B. N., Kudryavtseva, M. V., Sakuta, G. A. & Stein, G. I. Human hepatocyte polyploidization kinetics in the course of life cycle. Virchows Arch. B 64, 387–393 (1993). source
    • cites [15] Watanabe, T. & Tanaka, Y. Age-related alterations in the size of human hepatocytes. Virchows Arch. B 39, 9–20 (1982). source
    • cites [16] Saeter, G. et al. Changes in ploidy distributions in human liver carcinogenesis. J. Natl Cancer Inst. 80, 1480–1485 (1988). source

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

    Premisecitation

    Within the lobule, hepatocytes are classically organized into three zones, with zone 1 proximal to the portal vein and hepatic artery, zone 3 proximal to the central vein, and zone 2 in the intermediate region.

    “Within the lobule, hepatocytes are classically organized into three zones, with zone 1 proximal to the portal vein and hepatic artery, zone 3 proximal to the central vein, and zone 2 in the intermediate region3,4.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [3] Gebhardt, R. Metabolic zonation of the liver: Regulation and implications for liver function. Pharm. Therapeut 53, 275–354 (1992). source
    • cites [4] Jungermann, K. & Keitzmann, T. Zonation of Parenchymal and Nonparenchymal Metabolism in Liver. Annu Rev. Nutr. 16, 179–203 (1996). source

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

    Premisecitation

    Hepatocyte zonation matches function to the physiologic environment in both space and time.

    “Hepatocyte zonation matches function to the physiologic environment in both space and time5.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [5] Martini, T., Naef, F. & Tchorz, J. S. Spatiotemporal Metabolic Liver Zonation and Consequences on Pathophysiology. Annu. Rev. Pathol.: Mech. Dis. 18, 439–466 (2023). source

  • May combine two claims.

    Study design and conductown data

    Limited evidence of marker gene expression outside of the cell types they define was observed, supporting the segmentation.

    “Each cluster was defined by established markers, supporting our cluster assignment (Fig. 1b; Supplementary Data 2), and limited evidence of marker gene expression outside of the cell types they define was observed, supporting our segmentation (Supplementary Fig. 3c).”
    Quote in the paperSubstantial claim

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

    Premisecitation

    In adults, about a third of hepatocytes are multinucleated, or their nuclei contain more than two copies of each chromosome.

    “Early in life, most human hepatocytes are diploid and contain a single nucleus; ploidy increases with age such that in adults, about a third of hepatocytes are multinucleated, or their nuclei contain more than two copies of each chromosome14–16.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [14] Kudryavtsev, B. N., Kudryavtseva, M. V., Sakuta, G. A. & Stein, G. I. Human hepatocyte polyploidization kinetics in the course of life cycle. Virchows Arch. B 64, 387–393 (1993). source
    • cites [15] Watanabe, T. & Tanaka, Y. Age-related alterations in the size of human hepatocytes. Virchows Arch. B 39, 9–20 (1982). source
    • cites [16] Saeter, G. et al. Changes in ploidy distributions in human liver carcinogenesis. J. Natl Cancer Inst. 80, 1480–1485 (1988). source

  • Classified as probably not a substantial claim.

    Premisecitation

    The liver is composed of non-parenchymal cells, which include hepatic stellate cells (HSCs), resident macrophages, endothelial cells (ECs), and cholangiocytes.

    “The liver is also composed of non-parenchymal cells, which include hepatic stellate cells (HSCs), resident macrophages, endothelial cells (ECs), and cholangiocytes8,9.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [8] Damm, G. et al. Human parenchymal and non-parenchymal liver cell isolation, culture and characterization. Hepatol. Int. 7, 951–958 (2013). source
    • cites [9] Roberts, S., Ludwig, J. & Larusso, N. The pathobiology of biliary epithelia. Gastroenterology 112, 269–279 (1997). source

  • May combine two claims.

    Interpretationown data

    Hepatocyte zonation is at least partially maintained with fibrotic injury, while two injury-related hepatocyte states are enriched that do not appear to retain zonation.

    “Thus, we find that hepatocyte zonation is at least partially maintained with fibrotic injury, while two injury-related hepatocyte states are enriched that do not appear to retain zonation.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    These results suggest that multinucleated hepatocytes may contain more RNA transcripts within proportionally larger cells, which may underscore an increased metabolic capacity.

    “These results suggest that multinucleated hepatocytes may contain more RNA transcripts within proportionally larger cells, which may underscore an increased metabolic capacity.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The increase in RNA transcripts and cell size does not appear to be associated with differential gene expression or location of hepatocytes within the lobule, suggesting that these cells do not have distinct functional roles in the healthy liver, at least within the pathways explored with MERFISH.

    “However, this increase does not appear to be associated with differential gene expression or location of hepatocytes within the lobule, suggesting that these cells do not have distinct functional roles in the healthy liver, at least within the pathways we explored with MERFISH.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Hepatocytes have the capacity to replicate their nuclei without dividing, generating polyploid and multinucleated cells.

    “Hepatocytes have the capacity to replicate their nuclei without dividing, generating polyploid and multinucleated cells14–16.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [14] Kudryavtsev, B. N., Kudryavtseva, M. V., Sakuta, G. A. & Stein, G. I. Human hepatocyte polyploidization kinetics in the course of life cycle. Virchows Arch. B 64, 387–393 (1993). source
    • cites [15] Watanabe, T. & Tanaka, Y. Age-related alterations in the size of human hepatocytes. Virchows Arch. B 39, 9–20 (1982). source
    • cites [16] Saeter, G. et al. Changes in ploidy distributions in human liver carcinogenesis. J. Natl Cancer Inst. 80, 1480–1485 (1988). source

  • May combine two claims.

    Study design and conductown data

    The gene expression profiles of the snRNA-seq clusters identified with this analysis had expression profiles consistent with the corresponding MERFISH clusters, supporting their assignment, cross validating the expression profiles measured with both techniques, and providing further support for the MERFISH segmentation.

    “The gene expression profiles of the snRNA-seq clusters identified with this analysis had expression profiles consistent with the corresponding MERFISH clusters, supporting their assignment, cross validating the expression profiles measured with both techniques, and providing further support for the MERFISH segmentation (Supplementary Fig. 8b-d).”
    Quote in the paperSubstantial claim

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

    Study design and conductnone given

    All liver samples were collected from excess surgical tissue at Massachusetts General Hospital in accordance with protocols approved by the Mass General Brigham (MGB) or Dana Farber Cancer Institute (DFCI) Institutional Review Boards (IRBs).

    “All liver samples were collected from excess surgical tissue at Massachusetts General Hospital in accordance with protocols approved by the Mass General Brigham (MGB) or Dana Farber Cancer Institute (DFCI) Institutional Review Boards (IRBs) including approval to provide data on age by decade and sex.”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    No data that met the described criteria were excluded from the analyses.

    “No data that met the described criteria were excluded from the analyses.”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    The experiments were not randomized.

    “The experiments were not randomized.”
    Quote in the paperNot a substantial claim

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

    Study design and conductown data

    Cluster-level quality control using MALAT1 removed two clusters from the healthy dataset and two from the fibrotic injury dataset where >15% of cells contained fewer than three MALAT1 reads.

    “We removed two clusters from the healthy dataset and two from the fibrotic injury dataset where greater than 15% of cells contained fewer than three MALAT1 reads.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    All MERFISH and snRNA-seq data presented are from matched samples.

    “All MERFISH and snRNA-seq data presented within are from matched samples.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Cells with more nuclei are both larger and express higher levels of RNA.

    “These findings are consistent with the interpretation that cells with more nuclei are both larger15 and express higher levels of RNA.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [15] Watanabe, T. & Tanaka, Y. Age-related alterations in the size of human hepatocytes. Virchows Arch. B 39, 9–20 (1982). source

  • May combine two claims.

    Interpretationown data

    These results show the metabolic impact of chronic injury on hepatocytes, including common activation of nuclear hormone receptors and genes that regulate the urea pathway, and more restricted induction of transmembrane receptors, heat shock proteins, and the phospholipase PNPLA3.

    “These results show the metabolic impact of chronic injury on hepatocytes, including common activation of nuclear hormone receptors and genes that regulate the urea pathway, and more restricted induction of transmembrane receptors, heat shock proteins, and the phospholipase PNPLA3.”
    Quote in the paperSubstantial claim

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

    Study design and conductreasoning

    Sex was not considered as a variable due to the small sample size.

    “Sex was not considered as a variable due to the small sample size.”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    The Investigators were not blinded to allocation during experiments and outcome assessment.

    “The Investigators were not blinded to allocation during experiments and outcome assessment.”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    No statistical method was used to predetermine sample size for MERFISH or snRNA-seq.

    “No statistical method was used to predetermine sample size for MERFISH or snRNA-seq.”
    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

    Patients were consented under MGB protocol 2019P001245 (AM-061) or DFCI protocol 02-240.

    “Patients were consented under MGB protocol 2019P001245 (AM-061) or DFCI protocol 02-240 (all other samples).”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    CPS1 converts ammonia to carbamoyl phosphate to enter the urea cycle.

    “CPS1 (Carbamoyl phosphate synthetase I) is responsible for converting ammonia to carbamoyl phosphate to enter the urea cycle35”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [35] Martínez, A. I., Pérez-Arellano, I., Pekkala, S., Barcelona, B. & Cervera, J. Genetic, structural and biochemical basis of carbamoyl phosphate synthetase 1 deficiency. Mol. Genet. Metab. 101, 311–323 (2010). source

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

    Premisecitation

    SLC7A2 encodes an arginine/ornithine/lysine transporter that provides ornithine to interact with carbamoyl phosphate to form citrulline.

    “SCL7A2 encodes an arginine/ornithine/lysine transporter and provides ornithine (either directly or through conversion of arginine) to interact with carbamoyl phosphate to form citrulline36,37”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [36] Matsumoto, S. et al. Urea cycle disorders—update. J. Hum. Genet. 64, 833–847 (2019). source
    • cites [37] Yahyaoui, R. et al. A new metabolic disorder in human cationic amino acid transporter‐2 that mimics arginase 1 deficiency in newborn screening. J. Inherit. Metab. Dis. 42, 407–413 (2019). source

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

    Study design and conductown data

    The MERFISH barcoding scheme included 15 blank barcode controls not assigned to an RNA.

    “we leveraged the 15 not used to encode RNAs as ‘blank’ barcode controls, not assigned to an RNA (Supplementary Data 1).”
    Quote in the paperNot a substantial claim
94 of 94

  • The paper relies on this without support.

    The spatial organization of the human liver at single-cell resolution has not yet been described.

    “Single-cell RNA sequencing (scRNA-seq) has advanced our understanding of cell types and their heterogeneity within the human liver, but the spatial organization at single-cell resolution has not yet been described.”

  • The paper relies on this without support.

    The spatial organization of multinucleated hepatocytes and the impact of nuclear content on gene expression remain poorly understood.

    “Yet the spatial organization of multinucleated hepatocytes and the impact of nuclear content on gene expression remains poorly understood17.”

  • The paper relies on this without support.

    Many details of hepatocyte zonation, including how hepatocytes adjust gene expression along the portal-central vein axis and the function of this potential spatial fine tuning, remain unclear in the human liver.

    “However, many details of this zonation, including how hepatocytes adjust their gene expression along the portal-central vein axis and the function of this potential spatial fine tuning remain unclear in the human liver, as current single-cell analyses of the human liver have either measured gene expression without spatial data or evaluated spatial data without”

  • The paper relies on this without support.

    Subpopulations of macrophages and HSCs, and potential gene expression variations within, have not been resolved within the lobule organization at single-cell resolution in the human liver.

    “While sub-populations of macrophages and HSCs in the healthy human liver have been described in single-cell data10–12, these subpopulations, and potential gene expression variations within, have not been resolved within the lobule organization at single-cell resolution in the human liver.”

  • The paper relies on this without support.

    It is not clear if additional nuclear content leads to defined gene expression changes that could modulate function, or to what degree zonation influences nuclear content in the human liver.

    “It is not clear if additional nuclear content leads to defined gene expression changes that could modulate function, or to what degree zonation influences nuclear content in the human liver.”

  • The paper relies on this without support.

    Co-expression of receptor-ligand pairs in snRNA-seq data indicates potential cell-cell interactions in situ.

    “To explore this possibility, we leveraged CellPhoneDB31 with our snRNA-seq data to explore potential modes of cell-cell interactions that could be shaped by liver zonation (Methods; Supplementary Data 3).”

  • The paper relies on this without support.

    The fibrotic liver samples are representative of chronic liver injury and fibrosis.

    “In order to explore the spatial and cellular remodeling that occurs in the context of disease, we next performed MERFISH on human liver samples from patients with liver injury and fibrosis (Supplementary Table 1 and Supplementary Fig. 1).”

  • The paper relies on this without support.

    The liver samples analyzed are representative of healthy and fibrotic human liver.

    “profiling ~310,000 cells in both healthy and diseased liver”
8 of 8

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.

  • Proteinase K digestion duration differs

    Major

    The methods say samples were digested with proteinase K for two days, but the detailed protocol specifies a 24-hour incubation in the digestion solution.

1 of 1

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Watson, B. R., Paul, B., Rahman, R. U., Amir-Zilberstein, L., Segerstolpe, Å., Epstein, E. T., Murphy, S., Geistlinger, L., Lee, T., Shih, A., Deguine, J., Xavier, R. J., Moffitt, J. R., & Mullen, A. C.
Affiliations
  • Boston Children's Hospital (US)
  • University of Massachusetts Chan Medical School (US)
  • Broad Institute (US)
  • Massachusetts General Hospital (US)
  • Harvard University (US)
Corresponding author
  • Moffitt, Jeffrey R.
  • Mullen, Alan C.
ORCID
Venue
Nature Communications
Publisher
Springer Science and Business Media LLC
Published
2025-01-02
Volume
16
Issue
1
Pages
319-319
ISSN
2041-1723 (ISSN-L)
DOI
10.1038/s41467-024-55325-4
PMC
PMC11697218
OpenAlex
W4406016593
Semantic Scholar
40a7d19ca7efac7a59e9c929fab221b8cc653eed
Cited by
123
References
69
Citations by year
2026: 80 · 2025: 43
FWCI
30.97
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by-nc-nd)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • PubMed Central (submittedVersion, cc-by-nc-nd)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Chan Zuckerberg Initiative Seed Project
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics
  • Health Sciences › Medicine › Hepatology › Liver physiology and pathology
  • Health Sciences › Medicine › Epidemiology › Liver Disease Diagnosis and Treatment
Keywords
liver fibrosis, hepatocytes, hepatocyte ploidy, liver zonation, spatial transcriptomics, MERFISH, single-nucleus RNA sequencing, hepatic macrophages, mesenchymal cells, ligand-receptor interactions, liver injury, hepatocyte heterogeneity
MeSH terms
Adult, Female, Humans, Liver, Liver Cirrhosis, Macrophages, Male, In Situ Hybridization, Fluorescence, Sequence Analysis, RNA, Gene Expression Profiling, Hepatocytes, Single-Cell Analysis, Transcriptome

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 2026-10-22.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-22.

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