2025 年 72 巻 10 号 p. 1069-1078
Hepatic carbohydrate and lipid metabolism is strictly regulated by hormones such as insulin, glucagon, cortisol, and adrenaline, dynamically adapting to diet and stress. Metabolic zonation, a key feature of liver function, has been studied for decades. It refers to the spatial arrangement of hepatocytes with distinct metabolic roles along the portal-to-central vein axis, shaped by nutrient and oxygen gradients, as well as signaling molecules. However, traditional methods have struggled to reveal the spatial regulation of gene expression and signaling within these zones. Recent advances in single-cell and spatial omics technologies now allow detailed analysis of gene expression, signaling pathways, and cell-cell interactions with spatial resolution, providing new insights beyond classical models. Metabolic zonation research is rapidly advancing, and the concept of immune zonation, describing the spatial distribution of immune cells, has gained attention for its role in liver metabolism. These findings have improved our understanding of metabolic changes in conditions like fatty liver disease and diabetes. However, many questions remain, including the dynamic effects of diet and hormones and disease-related alterations. This review summarizes past and recent findings on metabolic zonation, explores the role of immune zonation and hormonal regulation, and discusses the latest technologies and future challenges.
The liver plays a key role in metabolic homeostasis by regulating gluconeogenesis and lipid synthesis. Hepatocytes have distinct metabolic functions depending on their location, a phenomenon known as metabolic zonation. Understanding this zonation is essential for studying liver function.
The concept of liver zonation dates back to the 1930s (Fig. 1). Smith and Kater showed that hepatocytes near the portal vein contain more mitochondria and glycogen than those near the central vein, demonstrating spatial differences in metabolism [1, 2]. In 1973, Rappaport proposed a three-zone model of the liver lobule, which became the foundation of the modern concept of metabolic zonation [3]. Jungermann and colleagues provided further evidence supporting functional differences among these zones [4-9].
Mall et al. first defined the hepatic lobule as the smallest liver unit in 1906. In 1933, Kater et al. demonstrated by electron microscopy that hepatocytes in the portal area contain a higher abundance of mitochondria. In 1973, Rappaport introduced the concept of zones based on gradients of nutrients, oxygen, and other factors, laying the foundation for the concept of metabolic zonation based on histological location. Subsequently, further insights were gained through immunostaining and other molecular biological techniques. More recently, the advent of single-cell and spatial omics has enabled zonation studies to advance in both higher resolution and novel perspectives. Created with Biorender.com.
In this model, blood flows from the portal vein to the central vein, creating gradients of oxygen and nutrients (e.g., glucose, amino acids, and lipids) that influence zonation. Recent studies have also revealed the role of Wnt signaling and transcription factors in regulating this process [10, 11]. Advances in single-cell and spatial omics technologies now allow more precise analysis beyond the Rappaport model, refining our understanding of liver zonation [12, 13].
Liver metabolic zonation is established and maintained by the spatial expression of enzymes, transcription factors, and signaling pathways, enabling each zone to perform distinct metabolic functions [12]. From the periportal region (Zone 1) to the pericentral region (Zone 3), each zone has specific roles. The molecular mechanisms underlying these functions are summarized below (Fig. 2).
Hepatocytes in each liver zone display distinct functional profiles. Zone 1 hepatocytes are primarily involved in gluconeogenesis, β-oxidation, and nitrogen metabolism, whereas Zone 3 hepatocytes predominantly engage in ketogenesis, triglyceride synthesis, glycolysis, and detoxification. Zone 2 serves as a transitional zone between these metabolic states. Each zone is characterized by a unique enzyme expression profile and is modulated by different transcription factors and signaling molecules. Key molecules include Phosphoenolpyruvate Carboxykinase 1 (PEPCK, PCK1), Carnitine Palmitoyltransferase 1 (CPT1, CPT1A), Carbamoyl-Phosphate Synthase 1 (CPS1), 3-Hydroxy-3-Methylglutaryl-CoA Synthase 2 (HMGCS2), Diacylglycerol O-Acyltransferase 1 (DGAT, DGAT1), Glycerol Kinase (GK), Cytochrome P450 Family 2 Subfamily E Member 1 (CYP2E1), Hepatocyte Nuclear Factor 4 Alpha (HNF4A), CCAAT Enhancer Binding Protein Alpha (CEBPA), Transcription Factor 7 Like 1 (TCF7L1), and T-Box Transcription Factor 3 (TBX3). Created with Biorender.com.
Zone 1 is primarily involved in gluconeogenesis, fatty acid β-oxidation, and nitrogen metabolism (urea cycle) [14, 15]. Enzymes such as pyruvate carboxylase and phosphoenolpyruvate carboxykinase (PEPCK) are highly expressed, supporting glucose production from non-carbohydrate precursors. Fatty acid β-oxidation is active, breaking down long-chain fatty acids to generate ATP. Additionally, nitrogen metabolism is regulated by high expression of carbamoyl phosphate synthetase 1 (CPS1) and ornithine transcarbamylase (OTC), two key enzymes in the urea cycle that detoxify ammonia and maintain nitrogen balance [16].
Zone 3 (Pericentral Region): Center of Glycolysis, Lipid Synthesis, and DetoxificationZone 3 is mainly responsible for energy production, lipid synthesis, and detoxification. Glycolysis is catalyzed by enzymes such as hexokinase and pyruvate kinase, ensuring ATP production to meet cellular energy demands. Unlike Zone 1, where fatty acid β-oxidation is dominant, Zone 3 favors fatty acid and cholesterol synthesis. Additionally, detoxification is a key function, with cytochrome P450 (CYP450) enzymes highly expressed, contributing to drug metabolism and reactive oxygen species clearance [5, 17, 18].
1.3. Regulatory Factors of Metabolic ZonationMetabolic zonation is strictly regulated by zone-specific transcription factors and signaling pathways. In Zone 1, transcription factors such as HNF4A and CEBPA regulate genes involved in gluconeogenesis and the urea cycle, supporting hepatocyte functions in glucose supply and nitrogen metabolism [19]. In contrast, in Zone 3, Wnt signaling through β-catenin regulates the expression of genes involved in lipid metabolism and drug detoxification, contributing to the metabolic specialization of Zone 3. Endothelial cells surrounding the central vein are key sources of Wnt ligands such as Wnt2, Wnt9b, and Rspo3, which shape the metabolic profile of Zone 3 [10, 11]. Additionally, transcription factors such as TCF7L1 and TBX3 suppress specific gene expression in Zone 3, playing a crucial role in maintaining metabolic zonation.
1.4. Bile Acid Signaling in ZonationBile acid metabolism is also regulated by zonation. In Zone 1, bile acid synthesis occurs through the classical pathway mediated by CYP7A1, while in Zone 3, the alternative pathway mediated by CYP27A1 is predominant, though species differences exist [20, 21]. Bile acid catabolism and excretion mainly occur in Zone 3, involving CYP3A4 and SULT2A1. Most secreted bile acids are reabsorbed in the intestine and returned to the liver via the portal vein, where they are primarily taken up in Zone 1. This process is regulated by FXR (Farnesoid X Receptor, NR1H4), which mediates feedback inhibition of bile acid synthesis [22]. Alterations in bile acid zonation have been implicated in diseases such as Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD), affecting disease progression [23].
1.5. Zonation MarkersIn single-cell analysis, zonation markers help determine the spatial localization of hepatocytes within the liver. These markers are closely linked to zone-specific metabolic functions and exhibit distinct gene expression patterns. In Zone 1, genes involved in gluconeogenesis (Pck1, G6pc) and the urea cycle (Cps1, Ass1) are highly expressed. Enzymes associated with fatty acid β-oxidation are also enriched in this region. In contrast, Zone 3 is characterized by high expression of genes such as Fasn and Scd1, which regulate lipid metabolism, and Glul, which is involved in ammonia detoxification. Zone 2 has been reported to express AXIN2 and LGR5, which are linked to stem cell function and tissue regeneration. Additionally, Zone 1 is associated with CDH1 (E-cadherin), a well-known marker for bile duct cells. In disease conditions such as MASLD, the expression of metabolism-related markers dynamically changes, necessitating careful data interpretation. For example, in our single-cell analysis of gluconeogenic pathways, Glul and Cdh1 were used as representative zonation markers instead of traditional gluconeogenesis markers [24]. The integration of spatial transcriptomics further improves the accuracy of zonation analysis by incorporating spatial information, allowing a more precise evaluation of liver zonation under both physiological and pathological conditions.
The relationship between metabolic zonation and the immune system has been highlighted as a key factor in liver function. The liver connects the portal and systemic circulation and serves as a defense barrier against pathogens entering through the portal vein from the gut [25]. The liver harbors an abundance of immune cells, including Kupffer cells, hepatic stellate cells, and a well-developed innate immune system consisting of Natural Killer (NK) cells and Natural Killer T (NKT) cells. These immune cells play a central role in immune responses by eliminating virus-infected and tumor cells (Fig. 3).
The liver is continuously exposed to pathogens via the portal vein and, as a result, possesses a well-developed innate immune system. Distinct zonal differences in immune cell localization, such as those observed in Kupffer cells, have been reported. Additionally, type 2 innate lymphoid cells (ILC2s) act on Zone 1 hepatocytes to suppress gluconeogenesis via IL-13 signaling. Moreover, the liver is enriched with monocyte-derived macrophages (MoMFs), natural killer/natural killer T cells (NK/NKT), and neutrophils, all of which are implicated in the inflammation observed in MASLD and MASH. Created with Biorender.com.
Although knowledge on the relationship between immune cells and liver zonation remains limited, Gola et al. demonstrated that bone marrow-derived and lymphoid cells, particularly Kupffer cells, are enriched in the periportal region (Zone 1), supporting to the concept of “immune zonation” [26]. Using quantitative multiplex imaging, genetic perturbation, transcriptomic analysis, infection models, and mathematical modeling, they showed that immune cells are preferentially localized in the periportal region. This asymmetric distribution was not due to developmental control but rather to MyD88-dependent signaling in liver sinusoidal endothelial cells (LSECs), triggered by commensal bacteria. This signaling regulated the composition of the pericellular matrix by forming chemokine gradients.
The interaction between the immune system and zonation may also influence glucose metabolism. Our study analyzed the relationship between innate immunity and hepatic gluconeogenesis using single-cell RNA sequencing and found that type 2 innate lymphoid cells (ILC2) suppress gluconeogenesis in periportal hepatocytes [24] (Fig. 3). Specifically, activation of ILC2 via IL-33 signaling suppressed hepatic gluconeogenesis, while this suppression was abolished in ILC2-deficient NSG mice. Furthermore, a CellPhoneDB-based cell-cell interaction analysis [27] suggested that this effect was mediated by IL-13 derived from ILC2, predominantly affecting periportal hepatocytes. Immunohistochemical analysis also confirmed that ILC2 was rarely detected in the pericentral region but was distributed from the periportal to the transitional zone. These findings suggest that immune cell localization may be closely related to metabolic zonation.
2.2. Immune Cells, Lipid Metabolism, and Disease ProgressionThe interaction between immune cells and metabolism also plays a crucial role in metabolic diseases such as MASLD [28]. In particular, the pericentral region (Zone 3) exhibits active fatty acid uptake and triglyceride synthesis, making it highly susceptible to lipid accumulation and inflammation [29]. While Kupffer cells are primarily localized in the periportal region, liver injury triggers to the recruitment of bone marrow-derived monocyte macrophages (MoMFs), which disrupt immune homeostasis by reducing Kupffer cell function, thereby exacerbating inflammation and fibrosis [30].
Liver macrophages are functionally diverse, consisting of Kupffer cells, bile duct-associated macrophages, and subcapsular macrophages. Even among Kupffer cells, distinct subtypes exist, with some involved in lipid metabolism and oxidative stress responses, while monocyte-derived Kupffer cells tend to induce inflammation and alter lipid metabolism. Changes in lipid metabolism can shift macrophage polarization and modify the immune microenvironment across different liver zones. While Kupffer cells are predominantly located in the periportal region and help maintain immune tolerance, monocyte-derived macrophages accumulate around the central vein in progressive Metabolic Dysfunction Associated Steatohepatitis (MASH), exacerbating disease pathology.
Thus, immune cell dynamics interact with liver metabolic zonation and may influence the progression of MASLD and fibrosis. The heterogeneous distribution of immune cells under physiological and pathological conditions is a key factor in understanding metabolic zonation.
The liver serves as both a source and a target of hormones (Fig. 4). However, the relationship between hormones and liver metabolic zonation remains largely unexplored. This section summarizes existing knowledge and discusses potential effects on different liver zones.
Liver metabolism is regulated by various hormones released from different organs. The liver itself releases several factors, such as Fibroblast Growth Factor 21 (FGF21), which can affect systemic metabolism.
Various hormone including thyroid hormone, cortisol, catecholamine, GH (growth hormone) and glucagon can regulate liver metabolism. Further research is needed to understand how hormones affect hepatocytes in different zones, particularly in the case of hormone-related disorders and liver diseases. Created with Biorender.com.
The liver produces FGF21, which regulates energy metabolism by promoting lipolysis, exerting anti-inflammatory effects, and improving insulin sensitivity in adipose tissue, skeletal muscle, and the central nervous system [31]. Additionally, the liver is involved in systemic hormone regulation through IGF-1 production (stimulated by growth hormone), conversion of T4 to T3, and steroid hormone metabolism [32, 33].
The pituitary-liver axis plays a crucial role in hormone secretion regulation. Growth hormone (GH), secreted by the pituitary gland, regulates energy homeostasis via hepatic IGF-1 production and metabolic control. In addition, the pituitary regulates cortisol secretion through adrenocorticotropic hormone (ACTH) and thyroid hormone secretion through thyroid-stimulating hormone (TSH). Consequently, pituitary dysfunction may directly impact liver metabolism. Qian et al. reported that the IRE1α-sXBP1 axis in the pituitary suppresses MASLD, but this pathway is impaired in obesity [34]. Furthermore, obesity reduces hepatic THRB (thyroid hormone receptor β) and XBP1 activity, suggesting a strong link between the pituitary-liver axis and liver disease progression.
Several key hormones influence hepatic glucose and lipid metabolism, including glucagon, GH, cortisol, catecholamines (adrenaline, noradrenaline), and thyroid hormones [35]. These hormones not only regulate physiological metabolism but also contribute to liver disorders such as fatty liver disease. For example, acromegaly is associated with increased gluconeogenesis and insulin resistance, glucagonoma causes hyperglycemia and lipid abnormalities, Cushing’s syndrome leads to hepatic fat accumulation and insulin resistance, and pheochromocytoma induces rapid blood glucose elevation and enhanced fatty acid mobilization. Additionally, thyroid dysfunction can lead to alterations in hepatic fat content and lipid abnormalities.
3.2. Hormones and Metabolic ZonationThe effects of hormones on hepatic glucose and lipid metabolism may vary across zones.
Thyroid hormones promote glycogen breakdown and gluconeogenesis, while also enhancing glycolysis and fatty acid oxidation, thereby facilitating overall metabolic balance. In lipid metabolism, they regulate cholesterol synthesis and fatty acid β-oxidation. However, an excess of thyroid hormones can result in hepatic fat reduction and hyperlipidemia [36]. While thyroid hormones influence the entire liver, their specific relationship with zonation remains unclear and requires further research.
Glucagon, secreted by pancreatic α-cells, strongly promotes gluconeogenesis and hepatic glycogen breakdown. Additionally, it enhances fatty acid β-oxidation and ketogenesis to ensure energy supply [37]. In terms of zonation, Zone 1 may exhibit increased expression of gluconeogenic enzymes (PEPCK, G6PC), whereas Zone 3 may favor fatty acid β-oxidation and ketogenesis.
GH increases the expression of gluconeogenic enzymes by inducing insulin resistance, helping to maintain blood glucose levels [35]. It also promotes fatty acid oxidation and mobilization from adipose tissue, optimizing energy utilization. Although GH has widespread effects, its zonal specificity remains unclear.
Cortisol, secreted by the adrenal cortex, is involved in energy supply during stress responses. It promotes gluconeogenesis and induces muscle protein breakdown to supply amino acids, while also enhancing glycogen synthesis for energy storage. In lipid metabolism, cortisol promotes fatty acid oxidation, but chronic excess can lead to hepatic steatosis and metabolic abnormalities [38]. These effects may be associated with gluconeogenesis in Zone 1 and lipid metabolism regulation in Zone 3.
Catecholamines, secreted by the adrenal medulla, rapidly induce metabolic changes in the liver during acute stress. They promote glycogen breakdown, leading to a rapid increase in blood glucose, while also activating fatty acid oxidation to enhance energy supply [39]. Adrenaline acts on the entire liver but may preferentially promote glycogen breakdown in Zone 1 and fatty acid oxidation in Zone 3.
Hormone distribution in the liver may be influenced by blood flow from the portal vein and hepatic artery. Glucagon is primarily delivered through the portal vein, whereas other hormones enter via the hepatic artery. Like nutrients and oxygen, hormones may establish concentration gradients across the liver lobule, but experimental validation remains challenging. Current evidence suggests that different hormones may have distinct effects on Zone 1 and Zone 3, highlighting the need for further research into their spatial regulation under both physiological and pathological conditions.
Recent advances in single-cell and spatial omics technologies have elucidated the molecular basis of liver zonation. This section provides an overview of these technologies and their applications. The liver consists of various cell types, including immune and endothelial cells, which actively influence metabolic zonation through cell-cell interactions (Fig. 5). Analytical tools such as CellPhoneDB [27] and CellChat [40] are widely used. CellPhoneDB predicts cell-cell interactions through a ligand-receptor database, while CellChat enables quantitative analysis of signaling networks. Furthermore, LIANA+ integrates spatial transcriptomics and metabolomics by incorporating multiple analytical tools (CellPhoneDBv2, CellChat, NATMI, Connectome, SingleCellSignalR, iTALK, CytoTalk) to reduce bias and enhance prediction accuracy [41].
Single cell RNA-seq has advanced the understanding of hepatocyte subpopulations across different zones and cell types. Spatial transcriptomics give us direct information of zone and transcriptomic data. With these technologies, we can now evaluate cell-cell interactions between hepatocytes and other cell types across different zones. A key limitation of spatial omics is its resolution, but this is steadily improving. Single-cell RNA-seq has the advantage of more precise cell annotation, so the combination of single cell RNA-seq and spatial transcriptomics is a promising approach. Transcription factor (TF) analysis and inferCNV can evaluate zone-specific TF activity and CNVs. The emerging long read sequencing technology could provide a new perspective for characterizing each liver zone. Created with Biorender.com.
Below, we present examples of applications in our research. The first example investigates the interaction between innate immunity and gluconeogenesis regulation. We applied CellPhoneDB to scRNA-seq data and demonstrated that interactions between periportal hepatocytes and ILC2 through IL-13/IL-13 receptor signaling contribute to gluconeogenesis suppression [24]. This allowed us to evaluate the heterogeneity of hepatocytes and innate immune cells at a single-cell level and provided insights into the relationship between innate immunity and liver metabolic zonation.
Another example involves analyzing the tumor microenvironment. We conducted scRNA-seq analysis on ten cases of craniopharyngioma and identified macrophage involvement specific to tumor subtypes using CellChat [42]. Additionally, transcription factor analysis with SCENIC identified key transcription factors in tumor and immune cells and mapped regulatory networks across cell states. SCENIC is a method that infers transcription factor-target gene relationships from gene expression data, making it useful for understanding pathways involved in tumor progression and immune responses [43]. Moreover, inferCNV [44] was used to evaluate copy number variants, revealing that CP9/10 (type 2 tumors) exhibited whole-cell copy number gains on chromosome 1q, whereas type 1 tumors did not. CP6/7/8 (mixed-type tumors) showed partial cell-specific copy number gains, suggesting genetic heterogeneity within tumor cells. Integrating these data with clinical and histopathological findings provided deeper insights into the transcriptional regulation and genetic heterogeneity of tumor cells, aiding in subtype classification and understanding tumor progression mechanisms. These methodologies are also valuable for elucidating liver zonation.
4.2. Advancements in Spatial Omics for Liver Zonation AnalysisGuilliams et al. combined CITE-seq, single-nucleus RNA-seq, spatial transcriptomics, and spatial proteomics to analyze macrophage niches in human and animal livers [45]. Their study demonstrated that lipid-associated macrophages (LAMs) accumulate in lipid-exposed regions and are induced by local lipid exposure. They also identified an evolutionarily conserved ALK1-BMP9/10 axis that regulates Kupffer cell and hepatic stellate cell crosstalk during liver development.
Currently, major spatial transcriptomics platforms include 10x Visium and Stereo-seq for whole-transcriptome analysis, while 10x Xenium is optimized for high-resolution targeted gene expression analysis. The limited resolution of spatial transcriptomics has been a major challenge for evaluating immune zonation. However, as whole-transcriptome platforms improve in resolution, more detailed analyses, including those involving the immune system, are becoming feasible. Bravo González-Blas et al. mapped the gene regulatory networks underlying liver zonation and identified TCF7L1 and TBX3 as key transcription factors in maintaining zonation [46]. Additionally, the deep learning model “DeepLiver” was used to predict enhancer activity, suggesting the possibility of artificially reproducing zonation patterns.
4.3. Future Technological DevelopmentsThis section explores emerging spatial omics technologies that may be applicable not only to liver studies but also to broader research areas. Matsumoto et al. provided a comprehensive review focusing on single-cell and spatial transcriptomics technologies, particularly in the pituitary gland, which may offer new insights [47].
4.3.1. Spatial MetabolomicsSpatial metabolomics is expected to become a key approach in liver metabolism research. Imaging Mass Spectrometry enables spatial analysis of metabolites in tissue slices. However, identifying specific cell types remains challenging, highlighting the need for integration with spatial transcriptomics [48]. This label-free technique allows simultaneous detection of multiple molecules and has potential applications in pathological tissue analysis and real-time disease progression monitoring [49].
4.3.2. Lipid ZonationLipid composition varies across liver zonation [50]. Stimulated Raman Scattering (SRS) microscopy, a label-free technique utilizing molecular vibrations (e.g., C-H and C=C bonds), enables differentiation of lipid types and structures. This method has been applied in various disease models, including brain disorders, while maintaining spatial resolution [51].
4.3.3. eQTL AnalysisExpression quantitative trait loci (eQTL) analysis identifies associations between genetic polymorphisms and gene expression, providing insights into disease-related SNPs. Liver eQTL studies have contributed to understanding metabolic disorders. A 2024 meta-analysis involving 1,183 liver tissue samples identified 9,013 eQTL signals across 6,564 genes. Integration with GWAS data for cardiometabolic traits revealed 1,582 colocalized GWAS-eQTL loci affecting 747 genes implicated in liver diseases [52].
In immunology, SNP effects are known to be cell-type- and environment-specific. Ishigaki et al. demonstrated that a specific SNP enhances CCR6 expression in B cells, increasing the risk of rheumatoid arthritis [53]. This cell-type-specific eQTL approach may be applicable to liver research, but reports on liver cell-type-specific eQTL analyses remain limited, particularly regarding metabolic zonation. Sakaue et al. integrated single-cell/single-nucleus multimodal RNA-seq with ATAC-seq to identify cell-type-specific enhancer activity and target genes associated with disease-related SNPs [54]. Currently, no single-cell eQTL studies focusing on liver zonation exist. However, future applications of this method may enable high-resolution analysis of SNP effects in periportal (Zone 1) and pericentral (Zone 3) regions. Single-cell eQTL analysis could uncover missing eQTLs undetectable in bulk analyses, leading to the identification of novel disease-risk variants. Future research focusing on zonation-specific gene regulation may enhance our understanding of the genetic basis of metabolic zonation and contribute to liver disease pathology and precision medicine.
4.3.4. Long-Read Sequencing (LRS)Long-read sequencing (LRS) technologies, such as single-molecule real-time (SMRT) sequencing (Pacific Biosciences) and nanopore sequencing (Oxford Nanopore Technologies), have emerged in recent years. Unlike conventional short-read-based scRNA-seq and spatial transcriptomics, LRS enables full-length transcript analysis, facilitating isoform expression evaluation and epigenetic modification analysis [55-57].
LRS is expected to advance metabolic zonation studies by analyzing molecules such as HNF4A P1/P2 isoforms, CYP2E1 splicing variants, and alternative splicing of GLUL (glutamine synthetase). Furthermore, LRS allows for the evaluation of epigenetic modifications, such as DNA methylation, which are difficult to detect with short-read sequencing. Integrating LRS with spatial transcriptomics may provide more detailed insights into transcriptional isoforms and epigenetic modifications in zonation.
The integration of single-cell transcriptomics, spatial transcriptomics, and metabolomics is being actively explored. This approach aims to link single-cell gene expression profiles with tissue-wide metabolic landscapes for a more comprehensive understanding of liver metabolic zonation. However, challenges remain, including the difficulty in assigning specific cell types to metabolomics data, and the need for normalization across omics scales. Overcoming these challenges will enable more precise multi-omics integration in the future.
This paper reviewed recent advancements in liver metabolic zonation research enabled by single-cell and spatial omics technologies (Graphical Abstract). It also discussed future directions, considering immune interactions and endocrinological aspects. Metabolic zonation is shaped by various factors, including dietary composition and hormonal signaling. Recent omics studies have provided new insights into cell-cell interactions, transcription factor networks, and the integration of multi-dimensional data, such as metabolomics. Despite these advancements, significant technical challenges remain, particularly in accurately delineating cellular boundaries in spatial omics and integrating multi-dimensional datasets across different platforms. Collaboration with researchers specializing in machine learning and data science will be essential to address these issues. Additionally, while large datasets, including those from human studies, continue to accumulate, and integration with genetic information such as SNPs is progressing, the relationship between hormones and metabolic zonation remains largely unexplored. This gap highlights an important area for future research. The integration of spatial omics technologies with clinical data, as presented in this paper, is expected to enhance our understanding of hepatic metabolic and endocrine disorders. This approach holds promise for identifying novel therapeutic targets and advancing personalized medicine.
This study was supported by the Japan Endocrine Society (JES) Grant for Promising Investigator. MF was also supported by grants from the Japan Society for the Promotion of Science (#22K08619, #22KK0271).
DisclosureTomoaki Tanaka is a member of Endocrine Journal’s Editorial Board. The author has no other competing interests to disclose.
AcknowledgementsWe would like to thank the members of the Tanaka Lab, as well as other lab members at Chiba University (K.Y., T.M., E.Y.L., S.S., and M.N.) and Harvard University (S.K., and M.L.) for their constructive discussions and suggestions on the manuscript.
https://ror.org/02f57qw79 
