The mammalian liver is a central hub for systemic metabolic homeostasis. Liver tissue is spatially structured, with hepatocytes operating in repeating lobules, and sub-lobule zones performing distinct functions. The liver is also subject to extensive temporal regulation, orchestrated by the interplay of the circadian clock, systemic signals and feeding rhythms. However, liver zonation has previously been analysed as a static phenomenon, and liver chronobiology has been analysed at tissue-level resolution. Here, we use single-cell RNA-seq to investigate the interplay between gene regulation in space and time. Using mixed-effect models of messenger RNA expression and smFISH validations, we find that many genes in the liver are both zonated and rhythmic, and most of them show multiplicative space-time effects. Such dually regulated genes cover not only key hepatic functions such as lipid, carbohydrate and amino acid metabolism, but also previously unassociated processes involving protein chaperones. Our data also suggest that rhythmic and localized expression of Wnt targets could be explained by rhythmically expressed Wnt ligands from non-parenchymal cells near the central vein. Core circadian clock genes are expressed in a non-zonated manner, indicating that the liver clock is robust to zonation. Together, our scRNA-seq analysis reveals how liver function is compartmentalized spatio-temporally at the sub-lobular scale.
Subscribe to Journal
Get full journal access for 1 year
only $9.92 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
All scRNA-seq data has been deposited in GEO with accession code GSE145197. Reconstructed spatio-temporal gene profiles are available as Matlab files at https://github.com/naef-lab/Circadian-zonation The whole dataset of gene profiles along with the analysis is available online as a web-application at the URL https://czviz.epfl.ch/. The application was built in Python using the library Dash by Plotly (version 1.0).
The code for fitting the mixed-effects models and generating the main figures is available at https://github.com/naef-lab/Circadian-zonation
Details regarding the statistics, software and data are provided in the Reporting Summary.
Gebhardt, R. Metabolic zonation of the liver: Regulation and implications for liver function. Pharmacol. Ther. 53, 275–354 (1992).
Hoehme, S. et al. Prediction and validation of cell alignment along microvessels as order principle to restore tissue architecture in liver regeneration. Proc. Natl Acad. Sci. USA 107, 10371–10376 (2010).
Ben-Moshe, S. & Itzkovitz, S. Spatial heterogeneity in the mammalian liver. Nat. Rev. Gastroenterol. Hepatol. 16, 395–410 (2019).
Colnot, S. & Perret, C. in Molecular Pathology of Liver Diseases Vol. 5 (ed. Monga, S. P. S.) 7–16 (Springer, 2011).
Wang, B., Zhao, L., Fish, M., Logan, C. Y. & Nusse, R. Self-renewing diploid Axin2+ cells fuel homeostatic renewal of the liver. Nature 524, 180–185 (2015).
Planas-Paz, L. et al. The RSPO–LGR4/5–ZNRF3/RNF43 module controls liver zonation and size. Nat. Cell Biol. 18, 467–479 (2016).
Jungermann, K. & Kietzmann, T. Zonation of parenchymal and nonparenchymal metabolism in liver. Annu. Rev. Nutr. 16, 179–203 (1996).
Halpern, K. B. et al. Single-cell spatial reconstruction reveals global division of labour in the mammalian liver. Nature 542, 352–356 (2017).
Dibner, C., Schibler, U. & Albrecht, U. The mammalian circadian timing system: organization and coordination of central and peripheral clocks. Annu. Rev. Physiol. 72, 517–549 (2010).
Green, C. B., Takahashi, J. S. & Bass, J. The meter of metabolism. Cell 134, 728–742 (2008).
Yeung, J. & Naef, F. Rhythms of the genome: circadian dynamics from chromatin topology, tissue-specific gene expression, to behavior. Trends Genet. 34, 915–926 (2018).
Maury, E., Ramsey, K. M. & Bass, J. Circadian rhythms and metabolic syndrome: from experimental genetics to human disease. Circ. Res. 106, 447–462 (2010).
Guan, D. et al. The hepatocyte clock and feeding control chronophysiology of multiple liver cell types. Science 369, 1388–1394 (2020).
Mermet, J., Yeung, J. & Naef, F. Systems chronobiology: global analysis of gene regulation in a 24-hour periodic world. Cold Spring Harb. Perspect. Biol. 9, a028720 (2017).
Zhang, R., Lahens, N. F., Ballance, H. I., Hughes, M. E. & Hogenesch, J. B. A circadian gene expression atlas in mammals: implications for biology and medicine. Proc. Natl Acad. Sci. USA 111, 16219–16224 (2014).
Sobel, J. A. et al. Transcriptional regulatory logic of the diurnal cycle in the mouse liver. PLoS Biol. 15, e2001069 (2017).
Vollmers, C. et al. Time of feeding and the intrinsic circadian clock drive rhythms in hepatic gene expression. Proc. Natl Acad. Sci. USA 106, 21453–21458 (2009).
Yeung, J. et al. Transcription factor activity rhythms and tissue-specific chromatin interactions explain circadian gene expression across organs. Genome Res. 28, 182–191 (2018).
Halpern, K. B. et al. Paired-cell sequencing enables spatial gene expression mapping of liver endothelial cells. Nat. Biotechnol. 36, 962–970 (2018).
Bunger, M. K. et al. Mop3 is an essential component of the master circadian pacemaker in mammals. Cell 103, 1009–1017 (2000).
Ripperger, J. A., Shearman, L. P., Reppert, S. M. & Schibler, U. CLOCK, an essential pacemaker component, controls expression of the circadian transcription factor DBP. Genes Dev. 14, 679–689 (2000).
Beal, J. Biochemical complexity drives log-normal variation in genetic expression. Eng. Biol. 1, 55–60 (2017).
McLean, R. A., Sanders, W. L. & Stroup, W. W. A unified approach to mixed linear models. Am. Stat. 45, 54 (1991).
Wagenmakers, E.-J. & Farrell, S. AIC model selection using Akaike weights. Psychon. Bull. Rev. 11, 192–196 (2004).
Atger, F. et al. Circadian and feeding rhythms differentially affect rhythmic mRNA transcription and translation in mouse liver. Proc. Natl Acad. Sci. USA 112, E6579–E6588 (2015).
Kietzmann, T. Metabolic zonation of the liver: the oxygen gradient revisited. Redox Biol. 11, 622–630 (2017).
Robles, M. S., Cox, J. & Mann, M. In-vivo quantitative proteomics reveals a key contribution of post-transcriptional mechanisms to the circadian regulation of liver metabolism. PLoS Genet. 10, e1004047 (2014).
Mauvoisin, D. et al. Circadian clock-dependent and -independent rhythmic proteomes implement distinct diurnal functions in mouse liver. Proc. Natl Acad. Sci. USA 111, 167–172 (2014).
Gebhardt, R. Liver zonation: novel aspects of its regulation and its impact on homeostasis. World J. Gastroenterol. 20, 8491 (2014).
Kornmann, B., Schaad, O., Bujard, H., Takahashi, J. S. & Schibler, U. System-driven and oscillator-dependent circadian transcription in mice with a conditionally active liver clock. PLoS Biol. 5, e34 (2007).
Reinke, H. et al. Differential display of DNA-binding proteins reveals heat-shock factor 1 as a circadian transcription factor. Genes Dev. 22, 331–345 (2008).
Sinturel, F. et al. Diurnal oscillations in liver mass and cell size accompany ribosome assembly cycles. Cell 169, 651–663.e14 (2017).
Kietzmann, T., Cornesse, Y., Brechtel, K., Modaressi, S. & Jungermann, K. Perivenous expression of the mRNA of the three hypoxia-inducible factor alpha-subunits, HIF1α, HIF2α and HIF3α, in rat liver. Biochem. J. 354, 531–537 (2001).
Ma, D., Panda, S. & Lin, J. D. Temporal orchestration of circadian autophagy rhythm by C/EBPβ: C/EBPβ regulates circadian autophagy rhythm. EMBO J. 30, 4642–4651 (2011).
Wang, J. et al. Nuclear proteomics uncovers diurnal regulatory landscapes in mouse liver. Cell Metab. 25, 102–117 (2017).
Harper, J. W., Ordureau, A. & Heo, J.-M. Building and decoding ubiquitin chains for mitophagy. Nat. Rev. Mol. Cell Biol. 19, 93–108 (2018).
Le Martelot, G. et al. REV–ERBα participates in circadian srebp signaling and bile acid homeostasis. PLoS Biol. 7, e1000181 (2009).
Ferrell, J. M. & Chiang, J. Y. L. Short-term circadian disruption impairs bile acid and lipid homeostasis in mice. Cell. Mol. Gastroenterol. Hepatol. 1, 664–677 (2015).
Ma, K. et al. Circadian dysregulation disrupts bile acid homeostasis. PLoS ONE 4, e6843 (2009).
Yu, L. et al. Disruption of Abcg5 and Abcg8 in mice reveals their crucial role in biliary cholesterol secretion. Proc. Natl Acad. Sci. USA 99, 16237–16242 (2002).
Oki, S. et al. ChIP‐Atlas: a data‐mining suite powered by full integration of public ChIP‐seq data. EMBO Rep. 19, e46255 (2018).
Gotic, I. et al. Temperature regulates splicing efficiency of the cold-inducible RNA-binding protein gene Cirbp. Genes Dev. 30, 2005–2017 (2016).
Jakobsson, A., Jörgensen, J. A. & Jacobsson, A. Differential regulation of fatty acid elongation enzymes in brown adipocytes implies a unique role for Elovl3 during increased fatty acid oxidation. Am. J. Physiol. Endocrinol. Metab. 289, E517–E526 (2005).
Burke, Z. D. & Tosh, D. The Wnt/β-catenin pathway: master regulator of liver zonation? Bioessays 28, 1072–1077 (2006).
Gougelet, A. et al. T-cell factor 4 and β-catenin chromatin occupancies pattern zonal liver metabolism in mice. Hepatology 59, 2344–2357 (2014).
Lavery, D. J. et al. Circadian expression of the steroid 15 α-hydroxylase (Cyp2a4) and coumarin 7-hydroxylase (Cyp2a5) genes in mouse liver is regulated by the PAR leucine zipper transcription factor DBP. Mol. Cell. Biol. 19, 6488–6499 (1999).
Carmon, K. S., Gong, X., Lin, Q., Thomas, A. & Liu, Q. R-spondins function as ligands of the orphan receptors LGR4 and LGR5 to regulate Wnt/beta-catenin signaling. Proc. Natl Acad. Sci. USA 108, 11452–11457 (2011).
Aizarani, N. et al. A human liver cell atlas reveals heterogeneity and epithelial progenitors. Nature 572, 199–204 (2019).
Brosch, M. et al. Epigenomic map of human liver reveals principles of zonated morphogenic and metabolic control. Nat. Commun. 9, 4150 (2018).
Mermet, J. et al. Clock-dependent chromatin topology modulates circadian transcription and behavior. Genes Dev. 32, 347–358 (2018).
Le Martelot, G. et al. Genome-Wide RNA polymerase II profiles and RNA accumulation reveal kinetics of transcription and associated epigenetic changes during diurnal cycles. PLoS Biol. 10, e1001442 (2012).
Wang, J. et al. Circadian clock-dependent and -independent posttranscriptional regulation underlies temporal mRNA accumulation in mouse liver. Proc. Natl Acad. Sci. USA 115, E1916–E1925 (2018).
Nitzan, M., Karaiskos, N., Friedman, N. & Rajewsky, N. Gene expression cartography. Nature 576, 132–137 (2019).
Ben-Moshe, S. et al. Spatial sorting enables comprehensive characterization of liver zonation. Nat. Metab. 1, 899–911 (2019).
Seglen, P. O. in Methods in Cell Biology Vol. 13, Ch. 4 (Elsevier, 1976).
Darnell, A. M., Subramaniam, A. R. & O’Shea, E. K. Translational control through differential ribosome pausing during amino acid limitation in mammalian cells. Mol. Cell 71, 229–243.e11 (2018).
Clayton, D. F., Harrelson, A. L. & Darnell, J. E. Dependence of liver-specific transcription on tissue organization. Mol. Cell. Biol. 5, 2623–2632 (1985).
Satija, R., Farrell, J. A., Gennert, D., Schier, A. F. & Regev, A. Spatial reconstruction of single-cell gene expression data. Nat. Biotechnol. 33, 495–502 (2015).
Loud, A. V. A quantitative stereological description of the ultrastructure of normal rat liver parenchymal cells. J. Cell Biol. 37, 27–46 (1968).
Kuhn, N. J., Woodworth-Gutai, M., Gross, K. W. & Held, W. A. Subfamilies of the mouse major urinary protein (MUP) multi-gene family: sequence analysis of cDNA clones and differential regulation in the liver. Nucleic Acids Res. 12, 6073–6090 (1984).
Hurst, J. L. et al. Individual recognition in mice mediated by major urinary proteins. Nature 414, 631–634 (2001).
Ilicic, T. et al. Classification of low quality cells from single-cell RNA-seq data. Genome Biol. 17, 29 (2016).
Maaten, Lvander & Hinton, G. Visualizing data using t-SNE. J. Mach. Learn. Res. 9, 2579–2605 (2008).
Müller, S., Scealy, J. L. & Welsh, A. H. Model selection in linear mixed models. Stat. Sci. 28, 135–167 (2013).
Schwarz, G. Estimating the dimension of a model. Ann. Stat. 6, 461–464 (1978).
Jammalamadaka, S. R. & SenGupta, A. Topics in Circular Statistics. Vol. 5 (World Scientific, 2001).
Chen, E. Y. et al. Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinf. 14, 128 (2013).
Lyubimova, A. et al. Single-molecule mRNA detection and counting in mammalian tissue. Nat. Protoc. 8, 1743–1758 (2013).
We thank D. Mauvoisin for assistance with the animal work, C. Gobet for bioinformatics advice and the EPFL BIOP facility for advice with microscopy. This work was supported by the Rothschild Caesarea Foundation’ fund managed by Weizmann Institute and EPFL, a Swiss National Science Foundation Grant 310030_173079 (to F.N.), and the EPFL. S.I. is supported by the Henry Chanoch Krenter Institute for Biomedical Imaging and Genomics, The Leir Charitable Foundations, Richard Jakubskind Laboratory of Systems Biology, Cymerman-Jakubskind Prize, The Lord Sieff of Brimpton Memorial Fund, the Wolfson Foundation SCG, the Wolfson Family Charitable Trust, Edmond de Rothschild Foundations, the I-CORE programme of the Planning and Budgeting Committee and the Israel Science Foundation (grants 1902/12 and 1796/12, the Israel Science Foundation grant no. 1486/16, the Chan Zuckerberg Initiative grant no. CZF2019-002434, the Broad Institute-Israel Science Foundation grant no. 2615/18, the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 768956), the Bert L. and N. Kuggie Vallee Foundation and the Howard Hughes Medical Institute (HHMI) international research scholar award.
The authors declare no competing interests.
Peer review information Primary Handling Editors: George Caputa; Pooja Jha. Nature Metabolism thanks Dominic Grun, Ueli Schibler and Jan S. Tchorz for their contribution to the peer review of this work.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
a, Histogram of number of UMIs per cell barcode for each mouse. Red patches mark the cells used for background estimation (100-300 UMI/cell barcode), gray patches mark the cells used for downstream analysis (1000-10000 UMI/cell barcode). b, Histogram of fraction of all UMIs mapping to mitochondrial genes. Filter used for downstream analysis in grey (0.09–0.35). c, smFISH staining of a liver lobule with probes against Cyp2e1 (red) and Cyp2f2 (green). CV = central vein, PN = portal node. Overall, the data combine 10 images from from two mice. d, Expression of Cyp2e1 and Cyp2f2 in cells with different fraction of mitochondrial expression. Three different filters for the fraction of UMIs mapping to mitochondrial genes (0–0.09, 0.09–0.35, 0.35–1) were applied, the data of all mice merged and the resulting datasets visualized as t-SNE plots. e, Violin plots for the correlations between Cyp2e1 and Cyp2f2 expression in single hepatocyte populations with different filters for fractions of mitochondrial expression. Each dot represents one mouse (n = 10 mice for each distribution) and the shape of the violin represents the density of points. f, Comparison of the zonation profiles of Z and Z + R genes obtained in our current study and the previous reconstruction from Halpern et al. 8. Profiles were interpolated to fit 15 layers, where 1 is pericentral and 8 is periportal. Dots indicate the center of mass (expression-weighted lobule layer) of the Z and Z+R genes computed in both datasets, for gene having an average expression of at least 10−5 in Halpern et al. r is the correlation coefficient, p the corresponding p-value from a standard linear regression.
Extended Data Fig. 2 Log-transformed reconstructed profiles, pre-filtering of the genes and comparison with external datasets.
a-c, Expression levels of the reconstructed profiles for the genes from Fig. 1f–h after log-transformation (Methods). Dots in represent data points from the individual mice. Lines represent mean expression per time point. Shaded areas represent one standard deviation (SD) across the mice (n=2 or n=3 depending on the time point, Methods). d, Biological variability of gene profiles across independent replicate liver samples, quantified in terms of the average relative replicate variance. 0 shows perfectly reproducible profiles while 1 the most variable genes (Methods). Genes inside the bottom-right box (x-cutoff at 10−5; y-cutoff at 0.5) are selected and contain all but one of the reference genes. Colored dots show reference zonated genes (blue) and reference rhythmic genes (orange). e, Comparison of the peak times for rhythmic genes in R and Z+R, with the dataset from Atger et al. PNAS 25. Circular correlation coefficient is 0.746 (Methods). f, Boxplot of the mRNA half-lives (data from Wang, J. et al. 35) shows that R genes as a group (median, orange line) are the shortest-lived. Box limits are lower and upper quartiles, whiskers extend up to the first datum greater/lower than the upper/lower quartile plus 1.5 times the interquartile range. Remaining points are marked. MW stands for the two-sided Mann-Whitney test, and KW stands for Kruskal-Wallis test.
Extended Data Fig. 3 Z+R and ZxR transcripts with corresponding rhythmic protein accumulation in bulk mass spectrometry data.
a-b, Rhythmic proteins corresponding to Z+R (a) and ZxR (b) transcripts were selected from Robles et al., 27 (from original Supplementary Table 2), and fitted with harmonic regression (p-value of rhythmicity from F-tests are indicated above the plot). Only proteins having a p-value<0.01 are shown. c, Scatter plot of the phase of the fits from the transcripts (x-axis) against the phase of the fits from the proteins (y-axis). The diagonal is indicated with a dashed grey line, the theoretical upper bound (6 h) for the delay between mRNA and protein is indicated with a dashed red line. All rhythmic proteins (q<0.2 in the original analysis) are represented.
a-b, smFISH (Stellaris, Methods) for Elovl3 (Z+R) and Arg1 (Z+R). smFISH quantifications were made for ZT0 and ZT12 (Methods). Left: representative images at ZT0, ZT12 for Elovl3 (a) or Arg1 (b). Pericentral veins (CV) and a periportal node (PN) are marked. Scale bar - 20 µm. Right: quantified profiles for each gene at the two time points from smFISH (top, line plot is the mean number of mRNAs, shaded area indicate SD across twelve images), and scRNA-seq data (bottom, line plot is mean expression, shaded area is SD across mice, n=2 or n=3 depending on the time point, Methods).
a, Spatial and temporal profiles and fits for circadian core-clock genes. Peak times are indicated on the temporal representation. For the genes Cry1 and Clock, additional dashed lines represent fits for the R model, as the Schwartz BIC weights from the R and Z+R models were close (Supplementary Table 2). b, Amplitudes and peak times of the core-clock circadian genes in a polar coordinate representation (clock-wise ZT times are indicated, distance from the center corresponds to the amplitude) show the expected organization of core clock transcript in the liver.
Extended Data Fig. 6 Spatio-temporal mRNA expression profiles for heat-responsive genes. Represented genes correspond to the top 200 targets from the ChIP-Atlas list for mouse HSF1.
a, Polar plot representation of the transcripts that are R, Z+R, or ZxR among the HSF1 targets from the ChIP-Atlas (http://chip-atlas.org/). Genes involved in chaperone functions (chaperones, co-chaperones, or chaperone facilitators) are named. Color indicates zonation, while grey dots show purely rhythmic genes. b, Spatial representation of the transcripts corresponding to proteins involved in chaperone functions, separated in central (left, cytoplasmic function) and portal (right, endoplasmic reticulum function) zonation.
Extended Data Fig. 7 Rhythmicity of Wnt targets in bulk RNA-seq, and proteomics liver time series data.
a, Enrichment of rhythmic genes (R, Z+R and ZxR) among the targets of the Wnt pathway, computed on the bulk dataset (Atger et al., 25). Targets above a given percentile (x-axis) of Apc-KO fold change are considered. The percentage of rhythmic genes in the whole Atger et al. dataset is indicated by a dashed blue line. b, Bulk mRNA (coming from Atger et al. dataset) rhythmicity profiles of Wnt targets among the top-50 targets with highest Apc-KO fold change. Gene profiles are centered around their mean. An enrichment of the phases around ZT8-14 is observed, in agreement with Fig. 6a. c, Polar plot representation of the individual gene phases and amplitudes represented in panel b (bulk data). d, Temporal representation of selected genes profiles from the scRNA-seq (top, n=2 or n=3 animals depending on the time point, Methods) and bulk proteomics (bottom, data from Robles et al. 27, n=2 replicates per time point sampled every 3 h) data. Represented profiles are ones with (1) the highest Apc-KO fold change, (2) a significantly rhythmic protein (p < 0.05, standard harmonic regression, F-test), and (3) belonging to the Z+R or ZxR category. e, Enrichment/depletion at different times (window size: 3 h), of both positive and negative Ras (N=31 and N=33, respectively) and Hypoxia (N=73 and N=41, respectively) targets (background: all R and Z+R genes). Colormap shows p-values (two-tailed hypergeometric test): red (blue) indicates enrichment (depletion).
About this article
Cite this article
Droin, C., Kholtei, J.E., Bahar Halpern, K. et al. Space-time logic of liver gene expression at sub-lobular scale. Nat Metab 3, 43–58 (2021). https://doi.org/10.1038/s42255-020-00323-1
Nature Metabolism (2021)
Nature Reviews Cancer (2021)