Abstract
2,3,7,8-Tetrachlorodibenzio-p-dioxin (TCDD) is a persistent environmental contaminant known for aryl hydrocarbon receptor (AHR)-mediated liver effects, including metabolic dysfunction-associated steatotic liver disease (MASLD)-like pathologies such as steatosis, inflammation, and fibrosis. Although previous studies have focused on AHR-mediated regulation of protein-coding genes, recent attention has turned to long non-coding RNAs (lncRNAs) because of their potential roles in the progression of steatotic liver disease (SLD). Using bulk and single-nuclei (sn)RNAseq datasets, we compared the dose-dependent AHR-mediated induction of lncRNA and mRNA expression by TCDD in the mouse and rat liver. This study also investigated cell-specific lncRNA-gene regulation within the murine liver to identify divergent lncRNA expression patterns across different hepatic cell types. Lastly, differentially expressed (DE) lncRNAs associated with human liver diseases were examined to investigate potential mechanistic roles. Comparative analysis of gene expression identified 2,386 mouse and 916 rat DE lncRNAs, with 203 common to both species. In contrast, mice had 6,071 compared to 3,056 rat DE mRNAs, with 1,492 in common. Integration of AHR genomic enrichment and putative dioxin response elements (pDRE) data with DE lncRNAs revealed regulation patterns similar to mRNA-coding genes, with both exhibiting greater frequency proximal to the transcription start site in both mice and rats. snRNAseq analysis also revealed 5,495 DE lncRNAs across all liver cell subtypes. Pericentral and periportal hepatocytes exhibited the most significant changes, with 3,339 and 3,550 DE lncRNAs respectively, followed by macrophages with 2,116. Among all DE genes, 52 previously annotated lncRNAs in hepatocytes were differentially expressed by TCDD, many of which are associated with steatosis, fibrosis, and hepatocellular carcinoma. Collectively, these results suggest AHR-mediated differential expression of lncRNAs may play a significant role in the progression of steatosis to steatohepatitis with fibrosis elicited by TCDD.
Introduction
Non-coding RNAs (ncRNAs) are transcribed RNA molecules that are not subsequently translated into proteins. Mammalian genomes are primarily comprised of non-coding regions, with up to 76–98% of the human genome encoding ncRNA sequences​ that vary in length, structure, and regulatory function1,2. Long non-coding RNAs (lncRNAs) are a subset of ncRNAs defined by their length of greater than 200 nucleotides and include antisense, intergenic (lincRNAs), intronic, enhancer RNAs (eRNAs), pseudogene-derived, and circular RNAs (circRNAs), with other ncRNAs identified with advances in sequencing technologies3,4. Although not translated into protein, lncRNAs serve pivotal roles in regulating gene expression, translation, chromatin structure, enzyme activity, cell differentiation and development, and can be differentially expressed by environmental stimuli4,5,6,7. Unlike messenger RNAs (mRNAs), which serve as templates for protein synthesis, lncRNAs exert effects through interactions with other biomolecules.
lncRNAs have emerged as key contributors in the progression and development of various diseases and disorders. These non-coding RNAs are reported to silence tumor suppressor genes, act as oncogenes, interact directly with transcription factors to enhance cell proliferation and survival, modulate epigenetic machinery, and promote cancer progression8. Beyond cancer, lncRNAs are linked to nephropathy9, cardiovascular disease10,11,12, rheumatoid arthritis13, Alzheimer’s disease, Huntington’s disease, and spinal muscular atrophy14 as well as associated with developmental disorders, including craniofacial defects4,15, brain and limb malformations16, impaired adipogenesis17, disrupted hematopoiesis18,19, male infertility20, and compromised liver function21. On a cellular level, lncRNAs are also involved in chromosome modification, nuclear transport and transcription activation10. In addition to these diverse roles and pathologies, lncRNAs regulate various metabolic pathways, including the metabolism of amino acids22, cholesterol23,24, glucose25,26, and lipids27. Consequently, dysregulated lncRNA expression may also contribute to the adverse effects elicited by environmental contaminants.
The aryl hydrocarbon receptor (AHR) is a ligand-activated transcription factor that regulates gene expression associated with various physiological and pathological processes, including xenobiotic metabolism, tissue development, immune response, and cellular proliferation28. AHR activation by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), and related compounds, leads to differential gene expression and subsequent pathobiological effects. Most, if not all, of these responses are AHR-mediated, as these effects are absent in Ahr-null mice29,30,31. In mice, TCDD elicits diverse liver pathologies including steatosis (lipid accumulation)32,33,34,35, inflammation33,34,35,36, and fibrosis33,36,37, reminiscent of metabolic dysfunction-associated steatotic liver disease (MASLD), a broad term used to characterize liver pathologies associated with metabolic dysfunction in humans38. More specifically, TCDD has been shown to disrupt hepatic lipid metabolism32,39,40,41, glycolysis42,43, amino acid metabolism41,44, the TCA cycle41, the urea cycle44, and nucleoside metabolism41. Although most studies examining the effects of TCDD have focused on the differential expression of protein-coding genes, a recent study reported the differential expression of more than 4,000 lncRNAs, affecting various cell types, transcriptional zonal distributions, and intercellular communication patterns related to fibrosis45. lncRNA dysregulation has also been implicated in metabolic-associated steatohepatitis (MASH), hepatocellular carcinoma (HCC), and cholestatic liver disease46.
Considering TCDD induces similar liver pathobiologies, we explored AHR-mediated dysregulation of lncRNA expression as a possible contributing factor in the dose-dependent progression of steatosis to steatohepatitis with fibrosis which are risk factors for more complex metabolic diseases including MASLD, diabetes, and hepatocellular carcinoma. Bulk RNA-sequencing (RNAseq) in mice and rats was re-assessed to determine species-specific differences in lncRNA expression following TCDD treatment, while single-nuclei RNA-sequencing (snRNAseq) in mouse liver was also re-examined for cell type- and zone-specific lncRNA expression. This included re-analysis of the same 28-day TCDD exposure snRNAseq dataset previously referenced45, which originally reported > 4,000 differentially expressed lncRNAs. Re-analysis uncovered dose-dependent, cell-specific lncRNA expression, with hepatocytes exhibiting distinct zonal lncRNA expression patterns. Integration with AHR ChIPseq and putative DRE (pDRE) data indicated that lncRNAs, like protein-coding mRNAs, were differentially expressed using canonical and non-canonical AHR pathways. Functional annotation of differentially expressed lncRNAs further implicated some in liver homeostasis and pathologies such as steatosis and fibrosis. These findings support a role for AHR-mediated lncRNA regulation in TCDD-elicited progression of steatosis and hepatotoxicity.
Methods
Datasets
Previously published in vivo studies were used to examine dose-dependent transcriptional responses to TCDD in male mice and female rats47,48. In these studies, postnatal day 25 (PND25) male C57BL/6Crl mice and female Sprague-Dawley rats were obtained from Charles River Laboratories. Animals were housed in Innovive Innocages under a 12-hour light/dark cycle with ad libitum access to Aquavive water and Harlan Teklad 22/5 Rodent Diet 8940. For both studies, a stock solution was prepared by suspending TCDD in sesame oil, as previously described39. Beginning on PND30, mice were orally gavaged with either 0.1 ml of sesame oil vehicle or 0.03, 0.1, 0.3, 1, 3, 10, or 30 µg/kg TCDD. Beginning on PND29, female rats were administered 0.1 ml of sesame oil vehicle or 0.01, 0.03, 0.1, 0.3, 1, 3, or 10 µg/kg TCDD. On day 28, four days after the last gavage, animals were euthanized by carbon dioxide overdose, tissues were collected, snap frozen, and stored at −80 °C. These data were chosen since female rats exhibit greater sensitivity to TCDD than their male counterparts, whereas in mice, males are more sensitive than females49,50. The dose ranges were adjusted between species to reflect these established differences in TCDD sensitivity48,51.
Gene models
A previously published custom Gene Transfer Format (GTF) annotation file that included 20,973 mRNAs and 55,038 lncRNAs52 was updated from the Mus musculus mm10 to mm39 genome build using UCSC liftOver (https://genome-store.ucsc.edu). The annotation file was also lifted over to the Rattus norvegicus rn7 genome build using UCSC liftOver to enable the identification of conserved novel lncRNAs. Post-liftover, the rn7 annotation file was filtered to retain only lncRNA annotations, resulting in 41,312 lncRNAs successfully transferred from the mouse genome. The rn7 annotation was subsequently augmented by incorporating 170 additional lncRNAs and 16,765 mRNA-coding genes annotated in the NCBI Rattus norvegicus reference.
Bulk RNAseq data processing
Previously published bulk RNAseq datasets (n = 5 per group) were used to investigate lncRNA expression changes following treatment with TCDD47,48. Quality assessment of the raw FASTQ files for mouse (GSE203302) and rat (GSE110293) was conducted using FASTQC v0.12.1 (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Adapter sequences were removed with Trimmomatic v0.3953. High-quality reads were then aligned to their respective reference genomes using STAR v2.7.154, with mouse reads mapped to the mm39 genome and rat reads to rn7. The custom GTF annotation files described above (mm39 for mouse and rn7 for rat) were used for read alignment and transcript quantification. Differential gene expression analyses were conducted using Deseq2 v1.34.055. For each species, vehicle-control samples were compared to TCDD-treated samples. Genes exhibiting a |fold-change| ≥ 1.5 and an adjusted p-value ≤ 0.05 were classified as differentially expressed.
snRNAseq data processing
Published single-nuclei RNAseq dose-response data (GSE148339) were downloaded from the Gene Expression Omnibus (GEO) repository56. This study assessed the hepatic effects of 0.01, 0.03, 0.1, 0.3, 1, 3, 10, or 30 µg/kg TCDD, or sesame oil vehicle, in mice orally gavaged every 4 days for a total of 28 days (7 treatments in total starting on day 0). Each treatment group included n = 2–3 biological replicates (independent mouse livers as previously described56. Data were generated using the 10X Genomics Chromium Single Cell 3ʹ v3.1 kit (10X Genomics, Pleasanton, CA). The raw FASTQ files were aligned to the mm39 reference genome using CellRanger v8.0.1 (10X Genomics). Aligned reads were annotated with a previously published custom GTF annotation file that included 20,973 mRNA-coding genes and 55,038 lncRNAs52. After alignment, reads were filtered to include only cells with: (i) a minimum of 100 genes detected, (ii) each gene detected in a minimum of 3 cells, and (iii) cells containing ≤ 1% mitochondrial DNA. Cell types were identified based on previously published annotations56,57 using ScanPy v1.10.058.
Differential gene expression analysis on single-cell RNAseq data yields more accurate results when performed on pseudobulk data, due to the inherent bias towards highly expressed genes59. For snRNAseq data, gene expression for each cell type at each dose was pseudobulked using Scater v1.22.060. Subsequent differential gene expression analyses were conducted using Deseq2 v1.34.055. Differential gene expression analysis was performed for each cell type by comparing vehicle-control to TCDD-treated samples. lncRNAs and mRNAs exhibiting a |fold-change| ≥ 1.5 and an adjusted p-value ≤ 0.05 were classified as differentially expressed.
pDRE identification
Putative dioxin response elements (pDREs) possessing the core sequence 5′-GCGTG-3′ were computationally identified and scored across both mm39 and rn7 genome builds using a previously published position weight matrix (PWM)61. Matrix similarity score (MSS) values were calculated independently within each species. Identical orthologous genes can yield different MSS scores due to sequence divergence in flanking bases. These thresholds were based on the minimum MSS values for functional (bona fide) AHR-bound DRE sites previously reported, which differ between species61. In the mm39 genome build, pDREs with a matrix similarity score (MSS) of ≥ 0.861 were included, while in rn7, a threshold of ≥ 0.825 MSS was applied. Datasets are available via the Harvard Dataverse (https://doi.org/10.7910/DVN/JASCVZ). Background frequencies of putative DREs across mammalian genomes, as well as comparisons to random sequence distributions, have been reported previously and provide a baseline for interpreting the enrichment patterns observed in the present study61.
AHR ChIPseq data processing
AHR genomic enrichment was determined in mice and rats using the 2-hour TCDD exposure datasets GSE97634 and GSE110282, respectively42,48. A previous analysis in mouse liver showed that AHR binding following TCDD treatment was highest at 2 hours, and yielded the greatest number of enriched peaks62. Therefore, ChIPseq data collected at 2-hours was utilized to provide the most inclusive assessment of potential regulatory events. AHR ChIPseq data (n = 5 per group; AHR and IgG antibodies) were originally aligned to older genome builds using Bowtie with peak calling performed in CisGenome. For the present analysis, genomic coordinates were updated to mm39 (mouse) and rn7 (rat) using liftOver, and peaks with FDR ≤ 0.05 were considered significantly enriched. These data were intersected with the custom annotation files described above to identify lncRNA genes with AHR genomic enrichment located within 10 kb upstream of the transcription start site (TSS) and extending through the 3’ end of the transcript.
K-means clustering
Clustering was performed using the K-means algorithm implemented in scikit-learn v0.24.263. The number of clusters was set to five to facilitate comparison to previously identified differentially expressed mRNAs. The K-means algorithm was applied to pseudobulk log2 fold-change values in cell types of interest, after which genes were assigned to clusters. Genes belonging to the same cluster were grouped for downstream analysis and visualization.
Results
Analysis of lncRNA differential expression in the liver elicited by TCDD was conducted using three gene expression datasets from male C57BL/6 mice and female Sprague-Dawley rats orally gavaged every 4 days for 28 days (Figure S1). The datasets included (i) snRNAseq from mouse liver, (ii) bulk RNAseq from mouse liver, and (iii) bulk RNAseq from rat liver. Previous histopathology assessments reported steatosis, immune cell infiltration, fibrosis and cell death using the same dose levels, treatment regimen and study design32,33,34,35,36. This included increased hepatocyte vacuolization at doses as low as 0.3 µg/kg, with necrosis and leukocyte infiltration at 3 µg/kg34. At 30 µg/kg, TCDD induced marked lipid accumulation, bile duct proliferation, inflammation, and fibrosis in male mice34. TCDD also modestly increased ALT levels and reduced body weight gain compared to controls (less than 15%) in the dose-response study, suggesting no overt toxicity.
Overview of AHR genomic enrichment
The distribution of pDREs and AHR genomic enrichment relative to the transcription start site (TSS) was assessed in both mouse (Fig. 1A and B) and rat genomes (Fig. 1C and D) to compare pDRE locations and AHR genomic enrichments relative to transcription start sites (TSS). In addition, expression of lncRNA (Fig. 1A and C) and mRNA (Fig. 1B and D) profiles for each species were separated. pDREs and AHR genomic enrichment frequency in mice for lncRNAs increased proximal to the TSS, comparable to mRNAs62. However, lncRNAs demonstrated a higher concentration of these elements upstream of the TSS, suggesting a differential regulatory landscape in these genomic regions with a similar assessment in the rat. Both lncRNAs and mRNAs exhibited increased frequencies of pDREs and AHR enrichment proximal to the TSS.
Genomic distribution of pDREs and AHR enrichment relative to transcription start sites (TSS) in mouse (mm39) and rat (rn7). Distribution of computationally identified putative Dioxin Response Elements (pDREs) and AHR genomic enrichment relative to the TSS for mouse (GSE97634) and rat (GSE110282) liver, two hours after oral gavage with TCDD. Data are shown for both lncRNA (left) and mRNA (right), with mouse (top) exposed to 30 µg/kg and rat (bottom) exposed to 10 µg/kg. Only pDREs and AHR-enriched regions within ± 10 kb of the TSS were depicted. Analyses were performed across all annotated genes with defined genomic coordinates (lncRNAs and mRNAs), and not restricted to only differentially expressed genes. Each data point represents the geometric center of a pDRE or AHR-enriched peak, binned across the ± 10 kb from the TSS. The y-axis shows the number of centers per bin.
Cross-species differential gene expression comparison
A previously-published gene model for the mouse genome consisting of 76,011 genes52 was used to identify mRNA and lncRNA transcripts. This transcript library was lifted over to the rat genome to assess potential transcript homology. Approximately 2,386 lncRNAs and 6,071 mRNAs were differentially expressed by TCDD in the mouse compared to 916 lncRNAs and 3,056 mRNAs in the rat, with both displaying dose-dependent responses liver (Fig. 2A-D). Dose-response gene expression patterns were clustered into five groups (Fig. 2A and B). Comparison of DE mouse and rat lncRNAs (Fig. 2C) and DE mRNAs (Fig. 2D) identified 203 DE lncRNAs and 1,492 DE mRNAs common to both datasets. Common DE transcripts exhibited induced and repressed lncRNA and mRNA clusters (Fig. 2E and F). A distinct characteristic observed solely in mouse lncRNAs was a cluster (denoted as a yellow cluster) where induction plateaued beginning at 1 µg/kg (Fig. 2E). These results underscore the similarity in the expression profiles of lncRNAs and mRNAs in both mouse and rat.
K-means clustering of differentially expressed genes shared between species. (A) Mouse and (B) rat bulk RNAseq data were assessed for dose-dependent differentially expressed (DE) lncRNAs and mRNAs. Heatmaps denote log2 fold-change (log2FC), with color scales indicating upregulation (red) or downregulation (blue). DE transcripts were clustered into 5 groups for each gene set. Sidebars indicate cluster assignments, with the corresponding average log2FC trends visualized in representative line graphs below each heatmap showing the gene expression profile for each cluster, with colors corresponding to the heatmap clusters. Venn diagrams show the number of differentially expressed (DE) lncRNAs (C) and mRNAs (D) shared transcripts between mouse and rat dose-response bulk RNAseq datasets. The overlapping regions represent genes that were commonly differentially expressed in both species. Overlapping DE (E) lncRNAs and (F) mRNAs were plotted for each species. Genes were considered differentially expressed if the |fold-change| ≥ 1.5 and the adjusted p-value ≤ 0.05. The assigned color to a cluster is specific to this figure and is not intended to correspond with clusters in other figures.
While there were common DE lncRNAs between mice and rats, notable differences were observed. Some lncRNAs exhibited induction or repression in both species (Fig. 3A and B), suggesting a conserved regulatory response to the treatment across species. However, others demonstrated divergent expression patterns, where one species showed dose-dependent induction and the other showed repression, indicating species-specific responses (Fig. 3C and D). These findings highlight the complexity of gene expression regulation across species and the importance of considering both conserved and species-specific responses.
Examples of differentially expressed lncRNAs with comparable and divergent expression patterns between species. Bar plots of differentially expressed (DE) lncRNAs in mouse (left) and rat (right). Bars denote log2fold-change (log2FC) of DE lncRNAs with (A,B) similar and (C,D) divergent expression profiles. Asterisk denotes differential expression (adjusted p-value ≤ 0.05).
Cross-cellular differential gene expression comparison
Dose-dependent differential gene expression was assessed using snRNAseq from male mice orally gavage with TCDD every 4 days for 28 days. Pericentral (PC) and periportal (PP) hepatocytes exhibited the highest number of DE lncRNAs, exhibiting 3,463 and 3,654 DE lncRNAs, respectively (Fig. 4A). In PC and PP hepatocytes, there were similar levels of DE lncRNAs that possessed comparable numbers of pDREs with AHR genomic enrichment. PC and PP hepatocytes also exhibited the highest number of DE mRNAs, with 5,372 and 5,722 genes, respectively (Fig. 4B). TCDD increased immune cell infiltration, with macrophages comprising ~ 30% of the cell population in the liver64. Macrophages exhibited 2,161 DE lncRNAs, of which approximately 9.9% (213) had pDREs within AHR-enriched genomic regions. In contrast, of the 5,890 DE mRNAs, 19.3% (1,135) had pDREs with AHR genomic enrichment. Other cell types with DE lncRNAs included liver sinusoidal endothelial cells (LSECs) and HSCs. Dose-dependent DE of lncRNAs and mRNAs was observed in all annotated cell types, with the majority of changes occurring at 10 and 30 µg/kg TCDD (Fig. 4C and D). In general, TCDD repressed lncRNA expression at the 30 µg/kg dose. mRNAs displayed comparable trends, with 10 and 30 eliciting the most DE mRNAs.
Dose-response expression profiles of differentially expressed lncRNAs and mRNAs across various cell types. Bars represent the total number of unique differentially expressed (DE) lncRNAs and mRNAs aggregated across all TCDD doses (0.01–30 µg/kg) in each cell type (black). Genes were considered differentially expressed if the |fold-change| ≥ 1.5 and the adjusted p-value ≤ 0.05. Differentially expressed (DE) lncRNAs and mRNAs were analyzed for the presence of AHR enrichment and pDREs. (A) lncRNAs and (B) mRNAs with at least one AHR-enriched region (dark gray), those with both AHR enrichment and at least one pDRE (light gray), and those with an AHR-enriched region that overlaps a pDRE (white) are shown. The number of DE lncRNAs and mRNAs with AHR enrichment centered around a pDRE is indicated in parentheses. The total number of induced (red) and repressed (blue) (C) lncRNAs and (D) mRNAs were determined for each dose in each cell type. Abbreviations: ECs (endothelial cells), HSCs (hepatic stellate cells), Macro (macrophages), PC Hep (pericentral hepatocytes), PFs (portal fibroblasts), PP Hep (periportal hepatocytes), Chol (cholangiocytes), pDCs (plasmacytoid dendritic cells), Neutro (neutrophils).
Given that hepatocytes represent ~ 70% of all cells in the liver, subsequent analyses focused on lncRNA expression in hepatocytes. K-means clustering divided hepatocyte DE lncRNAs and mRNAs into 5 distinct groups (Fig. 5). Both PP and PC hepatocytes exhibited dose-dependent induced and repressed DE lncRNA (Fig. 5A) and mRNAs (Fig. 5B) clusters, with lncRNAs showing a more gradual increase while mRNAs increased more sharply between 3 and 30 µg/kg. These findings further suggest similar regulation of lncRNAs and mRNAs, even between different cell types in the mouse liver.
K-means clustering of DE genes in periportal and pericentral hepatocytes. Differentially expressed (DE) (A) lncRNAs and (B) mRNAs in periportal (PP) and pericentral (PC) hepatocytes were clustered into 5 groups to identify dose-response trends. Heatmaps denote log2 fold-change (log2FC), with red and blue indicating upregulation and downregulation, respectively. Sidebar colors indicate cluster assignments. A representative expression profile is visualized in the line graphs below each heatmap showing the average log2FC for each cluster, with colors corresponding to the heatmap clusters. lncRNAs and mRNAs were considered differentially expressed if the |fold-change| ≥ 1.5 and the adjusted p-value ≤ 0.05. Cluster colors are specific to this figure and are not intended to correspond with clusters in other figures.
Zone-specific analysis of lncRNA expression
PP and PC hepatocytes perform distinct metabolic functions. To further investigate zonal differences, DE lncRNAs were compared between PP and PC subpopulations (Fig. 6). Between the 2,982 PP- and 3,077 PC-specific DE lncRNAs, 2,027 were expressed in both hepatocyte subtypes at 30 µg/kg TCDD (Fig. 6A). Most DE lncRNAs exhibited similar expression patterns regardless of the zone. For example, lnc757 was induced in PP and PC hepatocytes, while lnc34452 was repressed in both (Fig. 6B). However, exhibited opposing trends, such as lnc32381, which was inducted in PP hepatocytes but repressed in PC hepatocytes (Fig. 6C), while lnc9692 was repressed periportally but induced in pericentrally (Fig. 6C). In addition, some DE lncRNAs were zone specific, such as lnc5675 and lnc13556, which exhibited dose-dependent induction and repression, respectively, in PP hepatocytes, but were not changed pericentrally (Fig. 6D). Similarly, lnc13725 and lnc4345 were dose-dependently induced and repressed, respectively, in PC hepatocytes with no change in PP hepatocytes (Fig. 6E). Consequently, lncRNAs, like mRNAs, may exert zone-specific activities that contribute to the zone-specific effects of TCDD.
Differential expression (DE) of lncRNAs in periportal and pericentral hepatocytes. (A) The number of overlapping DE lncRNAs in periportal and pericentral hepatocytes. Numbers in parentheses indicate the DE lncRNAs unique to the cell type at 30 µg/kg. Zone-specific lncRNA expression profiles were investigated. Examples shown denote lncRNAs in which periportal and pericentral hepatocytes exhibited (B) similar or (C) divergent expression profiles, as well as expression changes exclusively in (D) periportal or (E) pericentral hepatocytes. Bars denote log2 fold-change (log2FC). Asterisk denotes differential expression (adjusted p-value ≤ 0.05). The percentage denotes the percentile rank of that lncRNA in the cell type, relative to all other genes expressed in that cell type. Abbreviations: PC (pericentral), PP (periportal).
lncRNAs and liver diseases
DE lncRNAs were annotated using data from the Mouse Genome Informatics (MGI) database to identify 48 differentially expressed known lncRNAs with an assigned official gene symbol (Fig. 7). Some DE lncRNAs were previously identified including Dleu2 and Bach2it1 that induce 4 and 3 lncRNAs, respectively. Others, including the Snhg family, showed PC and PP induction in response to TCDD. Dleu2, Peg13, and Ppp1r36dn were differentially expressed in PC hepatocytes while the induction of Pvt1, Lockd, and Hotair was specific to PP hepatocytes. Interestingly, Meg3 demonstrated divergent expression (induction in PP; repression in PC). Many of these annotated lncRNAs have been implicated in liver homeostasis or associated with tumor suppression, ferroptosis, fibrosis, cholangiocarcinoma, HCC, steatosis, and inflammation (Table 1).
MGI annotated lncRNAs differentially expressed in hepatocytes. Hepatocyte lncRNAs were annotated with data from the Mouse Genome Informatics (MGI) database65. lncRNAs were considered differentially expressed (*) when the |fold-change| ≥ 1.5 and the p-value ≤ 0.05. lncRNAs were ranked by transcript count abundance, with the most highly expressed transcripts shown in pink (100th percentile) and the lowest expressed transcripts in yellow (0th percentile). The presence of putative dioxin response elements (pDREs) and AHR genomic enrichment within the gene locus and 10 kb upstream of the transcription start site (TSS) is indicated for each gene (green). PC (pericentral), PP (periportal).
Discussion
TCDD induces the dose-dependent progression of steatosis to steatohepatitis with fibrosis and bile duct proliferation in male mice. This is accompanied by the disruption of multiple metabolic pathways essential for homeostasis32,34,35,36,37,39,42,43,44,47,115,116,117,118. Meta-analysis of 2,089 hepatic RNAseq datasets yielded a comprehensive list of hepatic lncRNA transcripts52. These annotations were used to examine the effects of TCDD on lncRNA hepatic expression in published dose-response snRNAseq datasets, and integrated with complementary pDRE, AHR ChIPseq, and histopathology data. The objectives of this study were to compare TCDD-elicited lncRNA and mRNA expression profiles and identify species-specific and conserved DE lncRNA and mRNA within mouse and rat datasets. In addition, cell type-specific expression patterns of lncRNAs across hepatic cell populations using snRNAseq data and potential associations between TCDD-responsive lncRNAs and known lncRNAs associated with human liver disease pathologies were examined.
Dere et al., reported that 43.2% of AHR-enriched genomic regions identified by genome-wide ChIP-chip analysis co-localize with a pDRE after 2 h of TCDD treatment when integrated with microarray differential gene expression data62. In the present study, pseudobulk analysis of snRNAseq data identified 3,969 DE mRNAs with AHR genomic enrichment that harbored at least one pDRE, representing 41.7% of all DE mRNAs, corroborating the Dere et al., study. However, of the 5,667 differentially expressed lncRNAs, only 1,131 (~ 20.0%) exhibited AHR genomic enrichment in regions containing at least one putative DRE following treatment with 30 µg/kg TCDD for 2 h. This suggests that while lncRNA expression can be regulated by AHR, it occurs to a lesser extent compared to mRNA-encoding genes. Canonically, AHR activation involves ligand binding (e.g., TCDD), nuclear translocation, dimerization with ARNT, and direct binding to DREs to regulate transcription119. However, non-canonical AHR signaling has also been reported where activation occurs independent of DREs, possibly through tethering to other transcription factors or chromatin-associated mechanisms119. The low incidence of pDREs among DE lncRNAs suggests AHR may regulate lncRNA expression primarily through non-canonical pathways or indirectly via secondary regulatory responses.
Despite the conservation in AHR structure and activation mechanism, TCDD-induced hepatotoxicity exhibits marked species-specific differences. Rats exhibit hepatocyte hypertrophy, while mice experience steatosis with inflammation and fibrosis51. In the present study, 1,492 mRNAs and 203 lncRNAs were identified as commonly differentially expressed in mice and rats. These shared DE genes exhibited similar expression patterns between species, clustering into similar profiles characterized by distinguishable dose-dependent induction or repression patterns. However, further analysis also found examples of divergent expression with most DE lncRNAs and mRNAs being unique to one species. For example, lnc19317 exhibited dose-dependent induction in mice but was repressed in rats, whereas lnc45640 showed the opposite trend, being repressed in mice and induced in rats. These divergent regulatory responses suggest differences in transcriptional networks activated by TCDD despite the conserved structure and activation mechanism of the AHR.
The liver lobule is organized into three distinct zones, with hepatocytes defined by their proximity to key vascular structures. Zone 1 (periportal) is adjacent to the portal triad, consisting of the portal vein, hepatic artery, and bile duct that deliver oxygen and nutrient rich blood to the lobule. Zone 3 (pericentral) surrounds the central vein, which collects deoxygenated blood and metabolic waste for drainage into the hepatic veins. Zone 2 bridges the periportal and pericentral zones, experiencing a gradient in oxygenation and metabolic activity throughout the liver lobule120. Each zone carries out distinct functions, with Zone 1 predominantly involved in oxygen-dependent metabolic processes, such as gluconeogenesis, fatty acid oxidation, and urea synthesis in both humans and mice. In contrast, Zone 3 is primarily engaged in glycolysis and detoxification processes, including the metabolism of xenobiotics and ammonia. Although humans and mice share a high degree of metabolic zone similarity, some pathways exhibit distinct zonation patterns between the two species. For example, multiple lipogenic genes are periportally located in humans but are pericentral in mice121. Enzymes responsible for alcohol and retinoid metabolism are pericentrally enriched in human liver but are periportal in the mouse122. TCDD has been reported to disrupt this zonal organization of mRNA-coding genes56,64,123.
In this study, the cell type-specific responses to TCDD, with a particular emphasis on lncRNA expression in PP and PC hepatocytes was examined. Of the 2,395 DE lncRNAs that were shared between periportal and pericentral hepatocytes, 1,259 and 1,068 DE lncRNAs were unique to PP or PC hepatocytes, respectively. lncRNAs, such as lnc757 and lnc34452, displayed similar dose-dependent expression in both, while others, such as lnc32381 and lnc9692, exhibited opposing expression responses. A subset also exhibited zone-restricted expression, such as lnc5675 and lnc13556 which were differentially expressed exclusively in PP hepatocytes, while lnc13725 and lnc4345 were regulated only in PC hepatocytes. This spatial regulation suggests that, like mRNA-encoding genes, lncRNAs may also contribute to the zonal specific effects of TCDD. However, liver-specific roles of lncRNAs are limited, compromising further elucidation of the contribution of AHR-mediated dysregulated of hepatic lncRNAs in TCDD elicited hepatoxicity.
Of the 3,463 PC and 3,654 PP DE lncRNAs, 51 were catalogued in the Mouse Genome Informatics (MGI) database and previously associated with liver pathologies. Malat1, Neat1, and Snhg3 have been implicated in hepatic steatosis, by modulating lipid metabolism through stabilization of SREBP-1c, altering H3K27 acetylation, and activating SND1/PPARγ signaling, respectively87,91,124. Altre inhibits hepatic inflammation70, while Airn, Malat1, Neat1, Platr4, and Trp53cor1 are associated with hepatic fibrosis by regulating LSEC differentiation, and modulating inflammatory chemokine expression like CXCL5, as well as p53-dependent apoptosis67,88,90,113. Airn, Bc1, Dancr, Dleu2, Dubr, Hotairm1, Hotairm2, Hoxaas2, and members of the Snhg family have been linked to hepatocellular carcinoma, whereas Jpx, Meg3, and Trp53cor1 exhibit tumor-suppressive activity68,72,73,74,75,76,77,79,80,81,82,83,84,86,89,93,95,96,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,114. These emerging functions highlight the under investigated role of lncRNAs in TCDD-elicited hepatotoxicity.
The AHR is well known for mediating the effects of environmental toxicants. AHR agonists dysregulated more than 1,400 lncRNAs, many of which can be linked to fatty acid metabolism, cell division, and immune response​125. A recent study in a human hepatocellular carcinoma model reported that AHR activation altered lncRNA expression associated with glucose and lipid metabolism, with ASAP1-IT1, RMDN2-AS1, and TP53TG5-6 identified as markers of poor prognosis, and FAM99B expression with improved outcomes126. In the present study, activation of AHR by TCDD induced Malat1 expression in PC hepatocytes with AHR genomic enrichment. Similarly, linc00673, which is associated with HCC progression, has induced in lung cancer mediated by the AHR125,127. These results suggest AHR-mediated lncRNA expression may influence hepatic disease progression.
Short-term gene expression assays have been proposed as an alternative to 2-year rodent studies to screen chemicals for carcinogenic potential128,129,130,131. These gene expression studies can distinguish genotoxic and non-genotoxic polyaromatic hydrocarbons as well as receptor-mediated toxicants132,133. However, these efforts have only considered mRNA-coding genes, potentially overlooking the mechanistic and biomarker significance of lncRNAs that could further improve the predictive sensitivity and accuracy of these screening approaches. Incorporating lncRNA analysis may also elucidate underlying mechanisms, providing a more comprehensive basis for suspected chemical toxicity and carcinogenicity.
In conclusion, this study demonstrates that lncRNAs are AHR regulated in a dose-, cell type-, and species-specific manner following TCDD exposure. The integration of single-nuclei RNAseq with AHR ChIPseq and pDRE mapping revealed that, while a lncRNA subset is regulated via canonical AHR-DRE interactions, the majority are differentially expressed through non-canonical mechanisms. Zone-specific lncRNA expression patterns in hepatocytes suggest spatially restricted responses to AHR activation, as reported with mRNA-coding genes. As with mRNAs, there were also species differences in lncRNA expression patterns in response to TCDD. Collectively, these findings expand the AHR regulatory network and suggest that lncRNAs may not only improve the predictive accuracy of toxicity and carcinogenicity gene expression screening assays but could also further elucidate the underlying mechanisms of toxicity of TCDD as well as other chemicals of concern.
Data availability
All RNAseq datasets used in this study have been previously published. Single-nuclei RNAseq dose-response data for mouse liver are available under accession number GSE14833956. Bulk RNAseq data for mouse and rat liver are available under GSE203302 and GSE110293, respectively47,48. ChIPseq datasets for AHR in mouse and rat are accessible via GSE97634 and GSE11028242,48. Computationally identified pDREs are available via Harvard Dataverse (https://doi.org/10.7910/DVN/JASCVZ). The code utilized in this study has been made accessible to enhance reproducibility (https://github.com/zacharewskilab/publication_analyses/tree/main/Analyses/TCDD_lncRNA_Repo).
References
Lee, H., Zhang, Z. & Krause, H. M. Long noncoding RNAs and repetitive elements: junk or intimate evolutionary partners? Trends Genet. 35(12), 892–902 (2019).
Mattick, J. S. Non-coding RNAs: the architects of eukaryotic complexity. EMBO Rep. 2(11), 986–991 (2001).
St Laurent, G., Wahlestedt, C. & Kapranov, P. The landscape of long noncoding RNA classification. Trends Genet. 31(5), 239–251 (2015).
Mattick, J. S. et al. Long non-coding RNAs: definitions, functions, challenges and recommendations. Nat. Rev. Mol. Cell Biol. 24(6), 430–447 (2023).
Dykes, I. M. & Emanueli, C. Transcriptional and post-transcriptional gene regulation by long non-coding RNA. Genom. Proteom. Bioinform. 15(3), 177–186 (2017).
Marchese, F. P., Raimondi, I. & Huarte, M. The multidimensional mechanisms of long noncoding RNA function. Genome Biol. 18(1), 206 (2017).
Zhang, X. et al. Mechanisms and functions of long non-coding RNAs at multiple regulatory levels. Int. J. Mol. Sci. 20(22), 5573 (2019).
Huarte, M. The emerging role of LncRNAs in cancer. Nat. Med. 21(11), 1253–1261 (2015).
Zhou, Q., Chen, W. & Yu, X-Q. Long non-coding RNAs as novel diagnostic and therapeutic targets in kidney disease. Chronic Dis. Translational Med. 5(4), 252–257 (2019).
Chen, Y., Li, Z., Chen, X. & Zhang, S. Long non-codingRNAs: from disease code to drug role. Acta Pharm. Sin B. 11(2), 340–354 (2021).
Zhang, Y., Du, W. & Yang, B. Long non-coding RNAs as new regulators of cardiac electrophysiology and arrhythmias: molecular mechanisms, therapeutic implications and challenges. Pharmacol. Ther. 203, 107389 (2019).
Yeh, C-F. et al. Expedition to the missing link: long noncoding RNAs in cardiovascular diseases. J. Biomed. Sci. 27(1), 48 (2020).
Zou, Y. & Xu, H. Involvement of long noncoding RNAs in the pathogenesis of autoimmune diseases. J. Transl Autoimmun. 3, 100044 (2020).
Sparber, P., Filatova, A., Khantemirova, M. & Skoblov, M. The role of long non-coding RNAs in the pathogenesis of hereditary diseases. BMC Med. Genomics. 12(Suppl 2), 42 (2019).
Long, H. K. et al. Loss of extreme long-range enhancers in human neural crest drives a craniofacial disorder. Cell. Stem Cell. 27(5), 765–783e714 (2020).
Sauvageau, M. et al. Multiple knockout mouse models reveal lincRNAs are required for life and brain development. Elife 2, e01749 (2013).
Sufianov, A. et al. The role of long non-coding RNAs in the development of adipose cells. Noncoding RNA Res. 8(2), 255–262 (2023).
Alvarez-Dominguez, J. R. et al. Global discovery of erythroid long noncoding RNAs reveals novel regulators of red cell maturation. Blood 123(4), 570–581 (2014).
Paralkar, V. R. et al. Lineage and species-specific long noncoding RNAs during erythro-megakaryocytic development. Blood 123(12), 1927–1937 (2014).
Lewandowski, J. P. et al. The Tug1 lcRNA locus is essential for male fertility. Genome Biol. 21(1), 237 (2020).
Mahpour, A. & Mullen, A. C. Our emerging understanding of the roles of long non-coding RNAs in normal liver function, disease, and malignancy. JHEP Rep. 3(1), 100177 (2021).
Hu, Q. et al. Amino acid metabolism regulated by lncRNAs: the propellant behind cancer metabolic reprogramming. Cell. Commun. Signal. 21(1), 87 (2023).
Hennessy, E. J. et al. The long noncoding RNA CHROME regulates cholesterol homeostasis in primate. Nat. Metab. 1(1), 98–110 (2019).
Ruan, X. et al. Identification of human long noncoding RNAs associated with nonalcoholic fatty liver disease and metabolic homeostasis. J. Clin. Investig. 131(1) (2021).
Zhang, P., Cao, L., Fan, P., Mei, Y. & Wu, M. LncRNA-MIF, a c-Myc-activated long non-coding RNA, suppresses glycolysis by promoting Fbxw7-mediated c-Myc degradation. EMBO Rep. 17(8), 1204–1220 (2016).
Zheng, X. et al. LncRNA wires up Hippo and Hedgehog signaling to reprogramme glucose metabolism. Embo j. 36(22), 3325–3335 (2017).
Duan, J. et al. Current advancements and future perspectives of long noncoding RNAs in lipid metabolism and signaling. J. Adv. Res. 48, 105–123 (2023).
Rejano-Gordillo, C. M. et al. From nucleus to organs: insights of aryl hydrocarbon receptor molecular mechanisms. Int. J. Mol. Sci. 23(23), 14919 (2022).
Boutros, P. C., Bielefeld, K. A., Pohjanvirta, R. & Harper, P. A. Dioxin-dependent and dioxin-independent gene batteries: comparison of liver and kidney in AHR-null mice. Toxicol. Sci. 112(1), 245–256 (2009).
Mimura, J. & Fujii-Kuriyama, Y. Functional role of AhR in the expression of toxic effects by TCDD. Biochim. Et Biophys. Acta (BBA) - Gen. Subj. 1619(3), 263–268 (2003).
Peters, J. M. et al. Amelioration of TCDD-induced teratogenesis in aryl hydrocarbon receptor (AhR)-null mice. Toxicol. Sci. 47(1), 86–92 (1999).
Nault, R., Fader, K. A., Lydic, T. A. & Zacharewski, T. R. Lipidomic evaluation of aryl hydrocarbon receptor-mediated hepatic steatosis in male and female mice elicited by 2,3,7,8-tetrachlorodibenzo-p-dioxin. Chem. Res. Toxicol. 30(4), 1060–1075 (2017).
Nault, R. et al. Dose-dependent metabolic reprogramming and differential gene expression in TCDD-elicited hepatic fibrosis. Toxicol. Sci. 154(2), 253–266 (2016).
Fader, K. A. et al. 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD)-elicited effects on bile acid homeostasis: alterations in biosynthesis, enterohepatic circulation, and microbial metabolism. Sci. Rep. 7(1), 5921 (2017).
Fader, K. A. et al. 2,3,7,8-Tetrachlorodibenzo-p-dioxin alters lipid metabolism and depletes immune cell populations in the jejunum of C57BL/6 mice. Toxicol. Sci. 148(2), 567–580 (2015).
Nault, R. et al. From the cover: Coagulation-driven hepatic fibrosis requires protease activated receptor-1 (PAR-1) in a mouse model of TCDD-elicited steatohepatitis. Toxicol. Sci. 154(2), 381–391 (2016).
Pierre, S. et al. Aryl hydrocarbon receptor-dependent induction of liver fibrosis by dioxin. Toxicol. Sci. 137(1), 114–124 (2014).
Rinella, M. E. et al. A multisociety Delphi consensus statement on new fatty liver disease nomenclature. J. Hepatol. 79(6), 1542–1556 (2023).
Cholico, G. N. et al. Thioesterase induction by 2,3,7,8-tetrachlorodibenzo-p-dioxin results in a futile cycle that inhibits hepatic beta-oxidation. Sci. Rep. 11(1), 15689 (2021).
Forgacs, A. L. et al. Comparative metabolomic and genomic analyses of TCDD-elicited metabolic disruption in mouse and rat liver. Toxicol. Sci. 125(1), 41–55 (2012).
Lin, S., Yang, Z., Liu, H. & Cai, Z. Metabolomic analysis of liver and skeletal muscle tissues in C57BL/6J and DBA/2J mice exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Mol. Biosyst. 7(6), 1956–1965 (2011).
Nault, R. et al. Pyruvate kinase isoform switching and hepatic metabolic reprogramming by the environmental contaminant 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Sci. 149(2), 358–371 (2016).
Cholico, G. N. et al. Consequences of reprogramming acetyl-CoA metabolism by 2,3,7,8-tetrachlorodibenzo-p-dioxin in the mouse liver. Sci. Rep. 13(1), 4138 (2023).
Cholico, G. N., Fling, R. R., Sink, W. J., Nault, R. & Zacharewski, T. Inhibition of the Urea cycle by the environmental contaminant 2,3,7,8-tetrachlorodibenzo-p-dioxin increases serum ammonia levels in mice. J. Biol. Chem. 105500 (2023).
Karri, K. & Waxman, D. J. TCDD dysregulation of LncRNA expression, liver zonation and intercellular communication across the liver lobule. Toxicol. Appl. Pharmacol. 471, 116550 (2023).
DiStefano, J. K. & Gerhard, G. S. Long noncoding RNAs and human liver disease. Annu. Rev. Pathol. 17, 1–21 (2022).
Orlowska, K. et al. Dioxin-elicited decrease in cobalamin redirects propionyl-CoA metabolism to the beta-oxidation-like pathway resulting in acrylyl-CoA conjugate buildup. J. Biol. Chem. 298(9), 102301 (2022).
Doskey, C. M. et al. 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) alters hepatic polyunsaturated fatty acid metabolism and eicosanoid biosynthesis in female Sprague-Dawley rats. Toxicol. Appl. Pharmacol. 398, 115034 (2020).
Pohjanvirta, R., Miettinen, H., Sankari, S., Hegde, N. & Linden, J. Unexpected gender difference in sensitivity to the acute toxicity of dioxin in mice. Toxicol. Appl. Pharmacol. 262(2), 167–176 (2012).
Pohjanvirta, R., Unkila, M. & Tuomisto, J. Comparative acute lethality of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), 1,2,3,7,8-pentachlorodibenzo-p-dioxin and 1,2,3,4,7,8-hexachlorodibenzo-p-dioxin in the most TCDD-susceptible and the most TCDD-resistant rat strain. Pharmacol. Toxicol. 73(1), 52–56 (1993).
Boverhof, D. R. et al. Comparative toxicogenomic analysis of the hepatotoxic effects of TCDD in Sprague Dawley rats and C57BL/6 mice. Toxicol. Sci. 94(2), 398–416 (2006).
Karri, K. & Waxman, D. J. Dysregulation of murine long noncoding single-cell transcriptome in nonalcoholic steatohepatitis and liver fibrosis. Rna 29(7), 977–1006 (2023).
Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for illumina sequence data. Bioinformatics 30(15), 2114–2120 (2014).
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29(1), 15–21 (2013).
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15(12), 550 (2014).
Nault, R. et al. Single-cell transcriptomics shows dose-dependent disruption of hepatic zonation by TCDD in mice. Toxicol. Sci. 191(1), 135–148 (2023).
Cholico, G. N., Nault, R. & Zacharewski, T. Cell-specific AHR-driven differential gene expression in the mouse liver cell following acute TCDD exposure. BMC Genom. 25(1), 809 (2024).
Wolf, F. A., Angerer, P. & Theis, F. J. SCANPY: large-scale single-cell gene expression data analysis. Genome Biol. 19(1), 15 (2018).
Squair, J. W. et al. Confronting false discoveries in single-cell differential expression. Nat. Commun. 12(1), 5692 (2021).
McCarthy, D. J., Campbell, K. R., Lun, A. T. L. & Wills, Q. F. Scater: pre-processing, quality control, normalization and visualization of single-cell RNA-seq data in R. Bioinformatics 33(8), 1179–1186 (2017).
Dere, E., Forgacs, A. L., Zacharewski, T. R. & Burgoon, L. D. Genome-wide computational analysis of dioxin response element location and distribution in the human, mouse, and rat genomes. Chem. Res. Toxicol. 24(4), 494–504 (2011).
Dere, E., Lo, R., Celius, T., Matthews, J. & Zacharewski, T. R. Integration of genome-wide computation DRE search, AhR ChIP-chip and gene expression analyses of TCDD-elicited responses in the mouse liver. BMC Genom. 12(1), 365 (2011).
Pedregosa, F. et al. Scikit-learn: machine learning in python. J. Mach. Learn. Res. 12(null), 2825–2830 (2011).
Nault, R., Fader, K. A., Bhattacharya, S. & Zacharewski, T. R. Single-nuclei RNA sequencing assessment of the hepatic effects of 2,3,7,8-Tetrachlorodibenzo-p-dioxin. Cell. Mol. Gastroenterol. Hepatol. 11(1), 147–159 (2021).
Baldarelli, R. M., Smith, C. L., Ringwald, M., Richardson, J. E. & Bult, C. J. Mouse genome informatics: an integrated knowledgebase system for the laboratory mouse. Genetics 227(1) (2024).
Shao, S. et al. LncRNA-Airn alleviates acute liver injury by inhibiting hepatocyte apoptosis via the NF-kappaB signaling pathway. Acta Biochim. Biophys. Sin (Shanghai). 54(11), 1619–1629 (2022).
Chen, T. et al. LncRNA Airn maintains LSEC differentiation to alleviate liver fibrosis via the KLF2-eNOS-sGC pathway. BMC Med. 20(1), 335 (2022).
Cai, H. et al. LncRNA AIRN influences the proliferation and apoptosis of hepatocellular carcinoma cells by regulating STAT1 ubiquitination. Arch. Pharm. Res. 44(4), 414–426 (2021).
Shen, S., Wang, J. & Lin, L. M. Downregulation of long non-coding RNA AIRN promotes mitophagy in alcoholic fatty hepatocytes by promoting ubiquitination of mTOR. Physiol. Res. 70(2), 245–253 (2021).
Ding, C. et al. A T(reg)-specific long noncoding RNA maintains immune-metabolic homeostasis in aging liver. Nat. Aging. 3(7), 813–828 (2023).
Ma, Y. & Cao, H. Lnc-ing’ T(reg) cells to the aging liver. Nat. Aging. 3(7), 760–761 (2023).
Ming, X. L. et al. Role of BCYRN1 in hepatocellular carcinoma pathogenesis by lncRNA-miRNA-mRNA network analysis and its diagnostic and prognostic value. Epigenomics 11(10), 1209–1231 (2019).
Ma, X. et al. DANCR acts as a diagnostic biomarker and promotes tumor growth and metastasis in hepatocellular carcinoma. Anticancer Res. 36(12), 6389–6398 (2016).
Guo, D. et al. DANCR promotes HCC progression and regulates EMT by sponging miR-27a-3p via ROCK1/LIMK1/COFILIN1 pathway. Cell. Prolif. 52(4), e12628 (2019).
Kim, S. S. et al. Early detection of hepatocellular carcinoma via liquid biopsy: panel of small extracellular vesicle-derived long noncoding RNAs identified as markers. Mol. Oncol. 15(10), 2715–2731 (2021).
Guo, Y. et al. LncRNA DLEU2 aggravates the progression of hepatocellular carcinoma through binding to EZH2. Biomed. Pharmacother. 118, 109272 (2019).
Liu, S. et al. SP1-induced lncRNA DUBR promotes stemness and oxaliplatin resistance of hepatocellular carcinoma via E2F1-CIP2A feedback. Cancer Lett. 528, 16–30 (2022).
Yang, L. & Jiang, J. GAS5 regulates RECK expression and inhibits invasion potential of HCC cells by sponging miR-135b. Biomed. Res. Int. 2019, 2973289 (2019).
Geng, Y. J., Xie, S. L., Li, Q., Ma, J. & Wang, G. Y. Large intervening non-coding RNA HOTAIR is associated with hepatocellular carcinoma progression. J. Int. Med. Res. 39(6), 2119–2128 (2011).
Wang, F. et al. HOX antisense lincRNA HOXA-AS2 promotes tumorigenesis of hepatocellular carcinoma. Cell. Physiol. Biochem. 40(1–2), 287–296 (2016).
Zhang, Y. et al. HOXA-AS2 promotes proliferation and induces epithelial-mesenchymal transition via the miR-520c-3p/GPC3 axis in hepatocellular carcinoma. Cell. Physiol. Biochem. 50(6), 2124–2138 (2018).
Lin, X. Q., Huang, Z. M., Chen, X., Wu, F. & Wu, W. XIST induced by JPX suppresses hepatocellular carcinoma by sponging miR-155-5p. Yonsei Med. J. 59(7), 816–826 (2018).
Ma, W. et al. Downregulation of long non-coding RNAs JPX and XIST is associated with the prognosis of hepatocellular carcinoma. Clin. Res. Hepatol. Gastroenterol. 41(2), 163–170 (2017).
Cagle, P., Qi, Q., Niture, S. & Kumar, D. KCNQ1OT1: an oncogenic long noncoding RNA. Biomolecules 11(11) (2021).
Sun, H., Li, Y., Kong, H., Dai, S. & Qian, H. Dysregulation of KCNQ1OT1 promotes cholangiocarcinoma progression via miR-140-5p/SOX4 axis. Arch. Biochem. Biophys. 658, 7–15 (2018).
Malakar, P. et al. Long noncoding RNA MALAT1 promotes hepatocellular carcinoma development by SRSF1 upregulation and mTOR activation. Cancer Res. 77(5), 1155–1167 (2017).
Yan, C., Chen, J. & Chen, N. Long noncoding RNA MALAT1 promotes hepatic steatosis and insulin resistance by increasing nuclear SREBP-1c protein stability. Sci. Rep. 6, 22640 (2016).
Leti, F. et al. Altered expression of MALAT1 LncRNA in nonalcoholic steatohepatitis fibrosis regulates CXCL5 in hepatic stellate cells. Transl Res. 190, 25–39e21 (2017).
He, Y. et al. Inhibitory effects of long noncoding RNA MEG3 on hepatic stellate cells activation and liver fibrogenesis. Biochim. Biophys. Acta. 1842(11), 2204–2215 (2014).
Huang, W., Huang, F., Zhang, R. & Luo, H. LncRNA Neat1 expedites the progression of liver fibrosis in mice through targeting miR-148a-3p and miR-22-3p to upregulate Cyth3. Cell. Cycle. 20(5–6), 490–507 (2021).
Hu, M. J., Long, M. & Dai, R. J. Acetylation of H3K27 activated LncRNA NEAT1 and promoted hepatic lipid accumulation in non-alcoholic fatty liver disease via regulating miR-212-5p/GRIA3. Mol. Cell. Biochem. 477(1), 191–203 (2022).
Zhang, Y., Luo, M., Cui, X., O’Connell, D. & Yang, Y. Long noncoding RNA NEAT1 promotes ferroptosis by modulating the miR-362-3p/MIOX axis as a CeRNA. Cell. Death Differ. 29(9), 1850–1863 (2022).
Bu, F. T. et al. LncRNA NEAT1: shedding light on mechanisms and opportunities in liver diseases. Liver Int. 40(11), 2612–2626 (2020).
Sun, Y. et al. Long noncoding RNA AI662270 promotes kidney fibrosis through enhancing METTL3-mediated m(6) A modification of CTGF mRNA. FASEB J. 37(8), e23071 (2023).
Wang, F. et al. Oncofetal long noncoding RNA PVT1 promotes proliferation and stem cell-like property of hepatocellular carcinoma cells by stabilizing NOP2. Hepatology 60(4), 1278–1290 (2014).
Zhang, S. & Song, X. Long non-coding RNA SNHG1 promotes cell proliferation and invasion of hepatocellular carcinoma by acting as a molecular sponge to modulate miR-195. Arch. Med. Sci. 16(2), 386–394 (2020).
Xie, X. et al. LncRNA Snhg3 aggravates hepatic steatosis via PPARγ signaling. Elife 13 (2024).
Zhao, Q. et al. LncRNA SNHG3 promotes hepatocellular tumorigenesis by targeting miR-326. Tohoku J. Exp. Med. 249(1), 43–56 (2019).
Qiu, J. et al. Long noncoding RNA SNHG4 promotes the malignant progression of hepatocellular carcinoma through the miR-211-5p/CREB5 axis. Cancer Med. 12(7), 8388–8402 (2023).
Li, Y. et al. LncRNA SNHG5 promotes the proliferation and cancer stem cell-like properties of HCC by regulating UPF1 and Wnt-signaling pathway. Cancer Gene Ther. 29(10), 1373–1383 (2022).
Chang, L. et al. Upregulation of SNHG6 regulates ZEB1 expression by competitively binding miR-101-3p and interacting with UPF1 in hepatocellular carcinoma. Cancer Lett. 383(2), 183–194 (2016).
Zhao, Z., Gao, J. & Huang, S. LncRNA SNHG7 promotes the HCC progression through miR-122-5p/FOXK2 axis. Dig. Dis. Sci. 67(3), 925–935 (2022).
Dong, J. et al. LncRNA SNHG8 promotes the tumorigenesis and metastasis by sponging miR-149-5p and predicts tumor recurrence in hepatocellular carcinoma. Cell. Physiol. Biochem. 51(5), 2262–2274 (2018).
Ye, S. & Ni, Y. LncRNA SNHG9 promotes cell proliferation, migration, and invasion in human hepatocellular carcinoma cells by increasing GSTP1 methylation, as revealed by CRISPR-dCas9. Front. Mol. Biosci. 8, 649976 (2021).
Lv, W. et al. Long non-coding RNA SNHG10 upregulates BIN1 to suppress the tumorigenesis and epithelial-mesenchymal transition of epithelial ovarian cancer via sponging miR-200a-3p. Cell. Death Discov. 8(1), 60 (2022).
Chen, P. P., Zhang, Z. S., Wu, J. C., Zheng, J. F. & Lin, F. LncRNA SNHG12 promotes proliferation and epithelial mesenchymal transition in hepatocellular carcinoma through targeting HEG1 via miR-516a-5p. Cell. Signal. 84, 109992 (2021).
Lin, R. X., Zhan, G. F., Wu, J. C., Fang, H. & Yang, S. L. LncRNA SNHG14 sponges miR-206 to affect proliferation, apoptosis, and metastasis of hepatocellular carcinoma cells by regulating SOX9. Dig. Dis. Sci. 67(3), 936–946 (2022).
Zhang, N., Lei, T., Xu, T., Zou, X. & Wang, Z. Long noncoding RNA SNHG15: A promising target in human cancers. Front. Oncol. 13, 1108564 (2023).
Liu, Q. et al. Long non-coding RNA SNHG16 as a potential biomarker in hepatocellular carcinoma: A meta-analysis. Med. (Baltim). 100(36), e27178 (2021).
Luo, Y. et al. LncRNA SNHG17 contributes to proliferation, migration, and poor prognosis of hepatocellular carcinoma. Can. J. Gastroenterol. Hepatol. 2021, 9990338 (2021).
Zhang, D., Cao, C., Liu, L. & Wu, D. Up-regulation of LncRNA SNHG20 predicts poor prognosis in hepatocellular carcinoma. J. Cancer. 7(5), 608–617 (2016).
Ye, Y. et al. Effect of lincRNA-p21 targeting HIF-1alpha on biological functions of liver cancer cells. Oncol. Lett. 17(6), 4964–4968 (2019).
Yu, F., Lu, Z., Chen, B., Dong, P. & Zheng, J. Identification of a novel lincRNA-p21-miR-181b-PTEN signaling cascade in liver fibrosis. Mediators Inflamm. 2016, 9856538 (2016).
Li, T. et al. Amplification of long noncoding RNA ZFAS1 promotes metastasis in hepatocellular carcinoma. Cancer Res. 75(15), 3181–3191 (2015).
Fling, R. R., Doskey, C. M., Fader, K. A., Nault, R. & Zacharewski, T. R. 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) dysregulates hepatic one carbon metabolism during the progression of steatosis to steatohepatitis with fibrosis in mice. Sci. Rep. 10(1), 14831 (2020).
Fader, K. A. et al. Convergence of hepcidin deficiency, systemic iron overloading, heme accumulation, and REV-ERBalpha/beta activation in aryl hydrocarbon receptor-elicited hepatotoxicity. Toxicol. Appl. Pharmacol. 321, 1–17 (2017).
Orlowska, K., Fling, R. R., Nault, R., Schilmiller, A. L. & Zacharewski, T. R. Cystine/glutamate Xc– antiporter induction compensates for transsulfuration pathway repression by 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) to ensure cysteine for hepatic glutathione biosynthesis. Chem. Res. Toxicol. 36(6), 900–915 (2023).
Angrish, M. M., Mets, B. D., Jones, A. D. & Zacharewski, T. R. Dietary fat is a lipid source in 2,3,7,8-tetrachlorodibenzo-rho-dioxin (TCDD)-elicited hepatic steatosis in C57BL/6 mice. Toxicol. Sci. 128(2), 377–386 (2012).
Wright, E. J., De Castro, K. P., Joshi, A. D. & Elferink, C. J. Canonical and non-canonical aryl hydrocarbon receptor signaling pathways. Curr. Opin. Toxicol. 2, 87–92 (2017).
Rappaport, A. M., Borowy, Z. J., Lougheed, W. M. & Lotto, W. N. Subdivision of hexagonal liver lobules into a structural and functional unit. Role in hepatic physiology and pathology. Anat. Rec. 119(1), 11–33 (2005).
Massalha, H. et al. A single cell atlas of the human liver tumor microenvironment. Mol. Syst. Biol. 16(12), e9682 (2020).
McEnerney, L. et al. Dual modulation of human hepatic zonation via canonical and non-canonical Wnt pathways. Exp. Mol. Med. 49(12), e413–e413 (2017).
Nault, R., Cholico, G. N. & Zacharewski, T. Analysis of cell-cell communication by single-nuclei RNA sequencing identifies AHR-mediated induction of NRG-ERBB signaling. Receptors 2(2), 148–159 (2023).
Orlowska, K. et al. Disruption of canonical AHR-mediated induction of hepatocyte PKM2 expression compromises antioxidant defenses and increases TCDD-induced hepatotoxicity. Redox Biol. 77, 103405 (2024).
Alluli, A., Fonseca, G., Matthews, J., Eidelman, D. H. & Baglole, C. J. Regulation of long non-coding RNA expression by aryl hydrocarbon receptor activation. Toxicol. Lett. 391, 13–25 (2024).
Xiao, X. et al. Aryl hydrocarbon receptor-regulated long non-coding rnas: implications for glycolipid metabolism and prognosis in hepatocellular carcinoma. Front. Oncol. 15 (2025).
Zhang, L. G., Zhou, X. K., Zhou, R. J., Lv, H. Z. & Li, W. P. Long non-coding RNA LINC00673 promotes hepatocellular carcinoma progression and metastasis through negatively regulating miR-205. Am. J. Cancer Res. 7(12), 2536–2544 (2017).
Corton, J. C. et al. A collaborative initiative to establish genomic biomarkers for assessing tumorigenic potential to reduce reliance on conventional rodent carcinogenicity studies. Toxicol. Sci. 188(1), 4–16 (2022).
Ganter, B. et al. Development of a large-scale chemogenomics database to improve drug candidate selection and to understand mechanisms of chemical toxicity and action. J. Biotechnol. 119(3), 219–244 (2005).
Igarashi, Y. et al. Open TG-GATEs: a large-scale toxicogenomics database. Nucleic Acids Res. 43(Database issue), D921–927 (2015).
Meier, M. J. et al. Progress in toxicogenomics to protect human health. Nat. Rev. Genet. (2024).
Ellinger-Ziegelbauer, H., Stuart, B., Wahle, B., Bomann, W. & Ahr, H. J. Characteristic expression profiles induced by genotoxic carcinogens in rat liver. Toxicol. Sci. 77(1), 19–34 (2004).
Corton, J. C. et al. A 50-gene biomarker identifies estrogen receptor-modulating chemicals in a microarray compendium. Chem. Biol. Interact. 394, 110952 (2024).
Funding
This project was supported by the National Institute of Environmental Health Sciences (NIEHS) Superfund Research Program [NIEHS SRP P42ES004911] and the NIEHS Research Project Grant Program [NIEHS R01ES029541] to T.Z. Partial support to T.Z. was provided by AgBioResearch at Michigan State University. G.N.C. was supported by NIEHS Multidisciplinary Training in Environmental Toxicology [T32ES007255].
Author information
Authors and Affiliations
Contributions
G.N.C. and T.Z. designed the project. G.A.P., G.N.C., and R.N. developed the transcriptomics workflow. G.A.P. and G.N.C. compiled all the data and produced the figures and table. G.A.P. and G.N.C. wrote the first draft of the manuscript. All authors reviewed and edited the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Post, G.A., Cholico, G.N., Nault, R. et al. Cell type- and species-specific regulation of hepatic lncRNAs by TCDD-activated aryl hydrocarbon receptor. Sci Rep 15, 37322 (2025). https://doi.org/10.1038/s41598-025-21357-z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-025-21357-z
