Hepatocyte HIF2α Downregulates the Urea Cycle Through Suppression of HNF4α Authors: Marc T Cantwell, Marwa O El-Derany, Raja Gopal Reddy Mooli, Sadeesh K Ramakrishnan, Yatrik M Shah Published: 2026 ABSTRACT AbstractBackground and AimsHypoxia contributes to chronic liver diseases such as metabolic dysfunction–associated steatotic liver disease through the activation of hypoxia-inducible factors (HIFs). Hif2α has been shown to drive a severe steatohepatitis and fibrosis phenotype in mice. How Hif2α drives this injury is not fully known. We identified a core regulatory pathway that was disrupted in hepatocytes in a Hif2α-dependent fashion to identify a possible common signaling pathway that may predispose to progression of inflammation and fibrosis.MethodsWe utilized mouse primary hepatocytes and mouse liver tissue from genetic models of chronic hypoxia along with standard molecular methods and ammonia assays to identify abnormal nitrogen metabolism.ResultsThe urea cycle is significantly depressed in a HIF2α-dependent fashion, which leads to elevation of ammonia. The urea cycle is controlled by a master regulator of hepatocyte function called hepatocyte nuclear factor 4 alpha (HNF4α), which is suppressed by Hif2α.ConclusionHif2α suppresses expression of a master regulator of hepatocyte biology, HNF4α, leading to significant disruption in hepatocyte-specific pathways such as the urea cycle. How Hif2α regulates HNF4α remains a point of active investigation, but inhibition of Hif2α may represent a therapeutic target to maintain the full function of a mature hepatocyte.Keywords: Hif2α, HNF4α, Urea Cycle Introduction Chronic liver disease (CLD) represents a continuum of pathological states ranging from minimal inflammation and fibrosis to advanced fibrosis and, ultimately, a terminal stage characterized by extensive fibrosis and regenerative nodules within an abnormally structured liver, termed cirrhosis. CLD also encompasses a broad spectrum of etiologies, including viral infections (eg, chronic hepatitis B and C), metabolic dysfunction (ie, metabolic dysfunction–associated steatotic liver disease [MASLD]), alcohol-associated liver disease, autoimmune liver diseases, and others.1 The relative prevalence of these etiologies fluctuates with epidemiologic trends and therapeutic advances. For example, the global obesity epidemic has led to a marked increase in MASLD, with an estimated one-third of the US adult population exhibiting at least hepatic steatosis.2 In contrast, direct-acting antiviral therapies have substantially reduced rates of cirrhosis and hepatocellular carcinoma secondary to hepatitis C virus infection.3 Although CLD arises from diverse etiologies, progressive fibrosis and architectural remodeling of the liver sinusoid ultimately disrupt the tightly regulated oxygen gradients within the hepatic acinus.4 The liver is unique in its dual blood supply from the portal vein and hepatic artery, which converge in the portal triads to deliver nutrient- and toxin-rich blood from the portal vein and oxygen-rich blood from the hepatic artery. As blood flows from the periportal region toward the central vein, oxygen tension predictably decreases.5 Disruption of this physiologic oxygen gradient induces hypoxia-inducible factor (HIF) signaling, which has been observed in both animal models and human samples of CLD, with increasing HIF expression in the hepatic parenchyma as fibrosis progresses to cirrhosis.6,7 HIF signaling is regulated through oxygen-sensitive prolyl hydroxylase domain (PHD) proteins.8 Under normoxic conditions, PHDs hydroxylate specific proline residues on the alpha subunits of HIF, of which there are 3 isoforms (HIF1α, HIF2α, and HIF3α). Hydroxylated HIFα subunits are recognized by the E3 ubiquitin ligase von Hippel-Lindau (VHL), leading to polyubiquitination and proteasomal degradation. Under hypoxic conditions, insufficient hydroxylation allows HIFα subunits to stabilize, dimerize with the constitutively expressed HIF1β subunit, translocate to the nucleus, and function as transcription factors to regulate gene expression.9 Prior work from our laboratory and others has demonstrated that HIF1α and HIF2α levels, along with their downstream effectors, increase during CLD progression in both animal models and human samples.10, 11, 12, 13 Previously, it was shown that constitutive activation of physiologic levels of HIF2α (but not HIF1α) drives rapid development of steatohepatitis and fibrosis in a rodent model.14 While some aspects of the metabolic dysregulation induced by chronic HIF2α activation in hepatocytes have been explored, further identification of aberrant metabolic pathways that promote fibrosis progression may reveal novel therapeutic targets. A central function of the liver is nitrogen metabolism, derived from dietary sources, gut microbial activity, and protein turnover in peripheral tissues such as muscle. The primary toxic byproduct of nitrogen metabolism is ammonia. As CLD progresses, urease-producing gut bacteria increase in abundance, generating higher ammonia loads that are delivered to the liver via the portal circulation for detoxification prior to systemic exposure.15 In a healthy liver, this ammonia is efficiently detoxified through the urea cycle without systemic consequences. The urea cycle comprises a series of enzymatic reactions occurring in both the mitochondria and cytoplasm that convert ammonia into the water-soluble waste product urea. With advancing liver disease, expression of key urea cycle enzymes, such as ornithine transcarbamylase (OTC), is reduced, leading to elevated systemic ammonia levels.16 A major clinical consequence of hyperammonemia is hepatic encephalopathy (HE), a neuropsychiatric syndrome characterized by altered cognition that frequently leads to hospital readmissions when severe enough to impair daily function. Beyond its role in HE, accumulating evidence suggests that elevated intrahepatic ammonia levels may be directly hepatotoxic, promoting inflammation and fibrosis and thereby establishing a positive feedback loop that accelerates progression to cirrhosis.17, 18, 19 Given the decline in urea cycle activity during CLD progression,20 the potential contribution of ammonia to hepatic inflammation and fibrosis, and the substantial clinical burden of chronic hyperammonemia, we sought to determine whether urea cycle dysfunction is regulated in a HIF-dependent manner. Identification of such a mechanism could reveal a targetable pathway to slow liver disease progression. Here, we demonstrate that ammonia levels are elevated in an animal model with physiologically relevant, constitutive activation of HIF signaling in hepatocytes and that this effect is specific to HIF2α rather than HIF1α. Furthermore, we show that this dysregulation is mediated by HIF2α-dependent downregulation of hepatocyte nuclear factor 4 alpha (HNF4α), a master transcriptional regulator of urea cycle enzyme expression. Methods Mouse Lines and Diets VhlF/F, VHLF/FHIF-2αF/F were previously described.14,21 For hepatocyte-specific disruption in a temporal manner (labeled Liv-VHL and Liv-VHL/HIF2α), these mice were crossed with cre-ERT2 recombinase mice under the control of the albumin promoter, SA-Cre-ERT2.22 Both male and female mice were utilized for these studies. These mice are a mixed Sv129 and C57BL/6 background, and their Cre-negative littermates were used as controls. The starting ages of the mice used were between 6 and 8 weeks for all experiments. To activate the SA-Cre-ERT2 recombinase, mice were treated with 75 mg/kg (approximately 2 mg) of tamoxifen dissolved in corn oil via intraperitoneal injection on 3 consecutive days and then harvested for either serum and liver tissue isolation or primary hepatocyte isolation 2 weeks after injection. The mice were housed in temperature- and light-controlled rooms (12-hour dark/light cycle) and were given water and pelleted chow ad libitum. For the ammonia serum and liver assays, wild-type mice at 8 weeks were placed on either the standard pelleted chow ad libitum or on a high-fat-containing diet consisting of 40% kcal by fat (mostly palm oil), 20% kcal by fructose, and 2% cholesterol for the time indicated for each treatment group (D09100310, Research Diets Inc, New Brunswick NJ). Mice were euthanized by inhalation of carbon dioxide and then by cervical dislocation to confirm euthanasia. All animals were housed at the Unit for Laboratory Animal Management at the University of Michigan. All animal studies were carried out in accordance with guidelines approved by the University of Michigan Animal Care and Use Committee. These studies were also conducted in accordance with ARRIVE 2.0 guidelines for the ethical and humane use of animals in research.23 Primary Hepatocyte Isolation and Cell Culture Primary hepatocytes were isolated from the mice as previously described.24 In brief, livers were perfused with 10 mL of Hank's Balanced Salt Solution containing collagenase type II (Worthington, Lakewood, NJ). Viable cells were plated in Dulbecco's Modified Eagle Medium media with 10% fetal bovine serum and 1% antibiotic-antimycotic. The cells were allowed to adhere overnight, followed by aspiration of old media, washing in phosphate buffered saline (PBS), and application of fresh media. The cells were then harvested 24 hours after isolation with lysis either in Trizol for RNA extraction or radioimmunoprecipitation buffer for protein extraction. Each mouse liver was considered a biological replicate, and the isolation was repeated on at least 3 different mouse livers. For the primary hepatocytes treated with hypoxia, HIF-2αFlF crossed to an albumin promoter driven cre-ERT2 recombinase were utilized. These mice are on a C57BL/6 background, and both male and female mice were utilized. Mice with the Cre recombinase and the littermate controls lacking the Cre recombinase were treated with tamoxifen for 3 days at 75 mg/kg via intraperitoneal injection and are labeled as wild type (WT) or knock out (KO) Liv-HIF2α. The primary hepatocytes were isolated 1 week after the last injection and plated. For the hypoxia treatment, they were placed into a 2% O2 chamber for 48 hours for the 48-hour time point; for the 24-hour time point they were incubated in normoxia for 24 hours before being placed in the hypoxia chamber for 24 hours. The control hepatocytes were incubated in normoxia for 48 hours; thus, all time points were harvested at a similar time point after isolation to control for any loss of mature markers from hepatocyte dedifferentiation. Serum and Liver Ammonia Assays Liv-VHL and Liv-VHL/HIF2⍺ were treated with tamoxifen as above and were sacrificed 2 weeks postinjection on a normal chow diet. For the diet assays, wild-type mice were placed either on diet for 16 weeks that consisted either of just normal chow, just the metabolic dysfunction–associated steatohepatitis (MASH) diet, or a combination of normal chow followed by the MASH diet for the indicated time (eg, MASH diet 4 weeks indicates 12 weeks of normal chow followed by 4 weeks of MASH diet). Whole blood (approximately 0.5 mL) was isolated from anesthetized mice via puncture of the v. temporalis superficialis in a terminal procedure in citrated tubes and placed on ice for 15 minutes. The whole blood was then spun down in a refrigerated centrifuge at 4 °C at 1000 g for 10 minutes. The supernatant was transferred to a new Eppendorf tube on ice, and then protein was precipitated utilizing 10% w/v trichloroacetic acid (Sigma-Aldrich, Saint Louis MO), utilizing 1 volume for 4 volumes of serum and spun down in a refrigerated centrifuge at 10,000 g for 10 minutes. The supernatant was aspirated, and 1 volume of PBS was added for trichloroacetic acid neutralization. The dilutions were factored into the final calculations for serum concentration. Ammonia was measured utilizing a fluorometric kit utilizing the provided standard curve per manufacturer instructions (Sigma-Aldrich, MAK310, St Louis, MO). One hundred mg of fresh liver tissue was briefly washed in cold 1x PBS and homogenized in an NP-40 lysis buffer containing 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid and protease inhibitor cocktail. This homogenate was spun down at 10,000 g for 15 minutes at 4 °C, and the supernatant was collected. Further protein extraction was conducted utilizing a chloroform/methanol mixture, and the protein-free methanol/water layer was isolated and used in the fluorometric assay; given the larger intrahepatic concentrations of ammonia, it was expressed as fold change compared to wild-type littermate control. Each mouse was considered a biological replicate, and at least 3 mice were utilized per treatment group. Gene Expression Analysis RNA was extracted from either homogenized whole livers or primary hepatocytes using the Trizol/chloroform method according to the manufacturer’s instructions. cDNA was synthesized from 1 μg (primary hepatocytes) or 2 μg (liver tissue) using the High-Capacity cDNA reverse transcription kit, and quantitative polymerase chain reaction) was performed per manufacturer instructions. Technical replicates were performed for each reaction in triplicate with at least 3 biological replicates (defined as one primary hepatocyte isolation or one mouse liver). Expression of all genes was normalized to beta-actin using the ΔΔCt method. The sequences of the primers for each target are found in the supplementary information section. Immunoblotting Lysates were prepared by homogenizing primary hepatocytes or PBS-washed liver tissue in radioimmunoprecipitation buffer (150 mM NaCl; 1% NP-40; 0.5% sodium deoxycholate; 0.1% sodium dodecyl sulfate; 50 mM Tris-HCl, pH 8.0) containing protease inhibitor cocktail, phenymethylsulfonyl fluoride, and sodium orthovanadate. The lysates were sonicated and precipitated by centrifugation at 20000x g for 10 minutes at 4 °C. Protein quantification was determined by bicinchoninic assay (Pierce). Lysates were resolved by sodium dodecyl sulfate-PAGE and transferred using standard procedures. Please see supplementary information for the primary antibodies utilized. Horseradish peroxidase-conjugated secondary antibodies were purchased from Cell Signaling and used at manufacturer-suggested dilutions. Chemi-blots were imaged utilizing iBright 1500 (Invitrogen), and densitometry of the blots was measured in Fiji and normalized to ß-Actin (Image J). Statistical Analysis Results are expressed as mean ± SEM. For analysis of 2 groups, Student’s t-test was utilized. For analysis of more than 2 groups, statistical significance was calculated by one-way analysis of variance followed by Tukey’s post hoc test or two-way analysis of variance followed by Sidak post hoc correction. P < .05 was considered statistically significant. GraphPad Prism 10 was used for data analysis. Results Ammonia Is Elevated in Both the Liver and the Serum of Liv-VHL KO Mice and Is Restored With HIF2α Deletion Given our prior work had demonstrated significant perturbation in lipid metabolism upon expression of physiologic levels of HIF2α, we analyzed other metabolic pathways that could be dysregulated in a HIF2α-dependent fashion.14 Through previous unpublished transcriptomic data, we identified significant downregulation in nitrogen-processing enzymes (data not shown). Given a metabolic pathway in the liver is the detoxification of ammonia via the urea cycle,25 we wanted to determine if ammonia levels are altered in the serum and livers following HIF activation in the hepatocytes. Genetic disruption of Vhl increases the expression of HIFs and HIF target genes. Liv-VHL WT and KO refer to the line with VhlF/F, either without Cre recombinase (WT) or with Cre recombinase (KO) driven by an albumin promoter, and Liv-VHL/HIF2 refer to the line with VhlF/FHif-2αF/F genes, with WT and KO denoting similar absence or presence of an albumin promoter driven Cre recombinase, respectively. Deletion of VHL in hepatocytes via tamoxifen injections significantly elevated serum ammonia levels compared to littermate control mice under normal chow conditions (Figure 1A). Consistent with these data, liver ammonia was also increased in Liv-VHL KO mice (Figure 1B). Given that deletion of Vhl leads to increased expression and activity of both HIF1α and HIF2α, as well as other possible pleiotropic effects from its deletion, we used the double knockout line of Vhl and Hif2 (Liv-VHL/HIF2α) to isolate the specific effects mediated by Hif2α. The Liv-VHL/HIF2 KO mouse showed normal levels of serum ammonia commensurate with the Liv-VHL WT mouse, indicating that despite the expected pleiotropic effects from Vhl deletion, it is specifically through the upregulation of Hif2α signaling that leads to elevated ammonia levels in both serum and liver (Figure 1A and B). As a positive control, it has been previously demonstrated that placing mice on a high-fat diet can also lead to increased levels of ammonia following chronic exposure to the diet (Figure 1A and B).19 Ammonia levels are elevated in mice in a HIF2⍺-dependent manner. (A) Serum of mice from either Liv-VHL or Liv-VHL/HIF2⍺ on a normal chow diet was taken 2 weeks after tamoxifen treatment; the diet arm consisted of wild-type mice fed either a normal chow/MASH diet for 16 weeks or normal chow followed by a MASH diet for the indicated time up to 16 weeks (eg, MASH 2 weeks is 14 weeks of normal chow followed by 2 weeks of MASH diet). (B) Liver tissue from the same animals in (A) expressed as a fold difference given the elevated concentrations compared to serum. N = 5 for each group, analysis of variance with Tukey’s for post hoc analysis. NS, not significant, ∗∗∗ = P < .001. Error bars = SEM. The Urea Cycle Is Downregulated in a HIF2α-Dependent Fashion Because the urea cycle accounts for the majority of hepatic ammonia detoxification,26 we next evaluated the integrity of this pathway in the Liv-VHL KO line. The rate-limiting step of the urea cycle is the mitochondrial conversion of ammonia to carbamoyl phosphate by carbamoyl phosphate synthetase I (CPS1), followed by condensation of carbamoyl phosphate with ornithine to form citrulline via OTC. Using primary hepatocytes, we quantified urea cycle enzyme expression by quantitative polymerase chain reaction in Liv-VHL WT and KO Liv-VHL mice and observed significant downregulation of all urea cycle genes, with the exception of the terminal enzyme arginase I (Figure 2A). This transcriptional suppression was also evident in whole-liver tissue (Figure 2B); however, in this context arginase I expression was significantly reduced as well. This difference likely reflects the distinct cellular composition of cultured hepatocytes compared with intact liver tissue, which contains both parenchymal and non-parenchymal cell populations, including sinusoidal endothelial cells, Kupffer cells, and infiltrating macrophages. Consistent with these transcriptional changes, immunoblot analysis of whole-liver lysates confirmed marked downregulation of OTC protein expression and a more modest reduction in CPS1 levels (Figure 2C). Although CPS1 catalyzes the rate-limiting step of the urea cycle, prior studies have similarly reported preferential suppression of OTC in CLD. Importantly, loss of OTC activity has well-established clinical relevance, as patients with OTC deficiency develop severe hyperammonemia when exposed to high protein loads, leading to neurotoxicity, coma, and potentially death.27 These observations underscore the functional significance of OTC downregulation in the context of CLD-associated hyperammonemia. Importantly, suppression of the urea cycle was dependent on HIF2α signaling. In contrast to Liv-VHL KO mice, double-knockout Liv-VHL/HIF2α KO animals exhibited urea cycle gene expression in the whole liver that was indistinguishable from WT controls, along with normalization of OTC protein levels (Figure 3A and B). Together, these data identify HIF2α as a critical mediator of urea cycle repression in this model. Urea cycle enzymes are downregulated in VHL knockout mice. (A) Primary mouse hepatocytes were isolated, and quantitative polymerase chain reaction performed for urea cycle enzymes, each N being an individual mouse that served as the source for primary hepatocytes (N = 6). In (B) whole livers were analyzed for the expression of the urea cycle components via quantitative polymerase chain reaction (N = 9). In (C), a western blot of the 2 important regulatory enzymes of the urea cycle from isolated mouse livers with their quantitation (N = 3). Quantitative polymerase chain reaction analyzed via analysis of variance, ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, ∗∗∗∗P < .0001. Error bars = SEM. Urea cycle expression is HIF2⍺ dependent. (A) Livers from the double-knockout mice Liv-VHL/HIF2⍺ show normal expression of urea cycle enzymes (N = 6). (B) Liver OTC protein levels are normal in double-knockout mice. Error bars = SEM. HIF2α Downregulates Hepatocyte Nuclear Factor 4 Alpha, the Major Transcriptional Regulator of the Urea Cycle We next sought to define the mechanism by which HIF2α regulates the urea cycle. Prior studies have established that urea cycle gene expression is governed largely by HNF4α, a master transcriptional regulator of hepatocyte differentiation and function.28 HNF4α is indispensable for liver development, as hepatocyte-specific deletion of HNF4α results in embryonic lethality.29 Beyond its developmental role, HNF4α is a central regulator of hepatic metabolic programs, including gluconeogenesis, ketogenesis, fatty acid metabolism, and xenobiotic metabolism.30,31 HNF4α expression is reduced in MASH and in experimental high-fat diet models. In these settings, decreased HNF4α levels lead to derepression of p53-dependent pathways, thereby increasing hepatocyte sensitivity to lipotoxicity and apoptosis.32, 33, 34 Notably, restoration of HNF4α expression via mRNA delivery in both mouse and human primary hepatocytes, as well as in vivo mouse models, has been shown to re-establish key metabolic pathways in fibrotic hepatocytes and to confer protection against toxin- and cholestasis-induced fibrosis.35 Given the role of HNF4α as a master regulator of hepatocyte function and its emerging potential as a therapeutic target,36 we next examined whether HIF2α regulates HNF4α expression or activity. HNF4α mRNA levels were significantly reduced in Liv-VHL knockout livers, whereas expression was preserved in double-knockout Liv-VHL/HIF2α livers (Figure 4A). Consistent with these findings, HNF4α protein levels were also decreased in Liv-VHL knockout livers. Together, these data position HIF2α upstream of HNF4α and indicate that hypoxia signaling through HIF2α disrupts a central transcriptional regulator of hepatocyte identity and metabolic function. HNF4α expression is decreased in a HIF2α manner. (A) Livers from the Vhl KO mouse have decreased expression of HNF4α, which is restored when HIF2α is also deleted (N = 5–10). (B) Western blot from Vhl KO mice demonstrating decreased protein levels in line with the decreased mRNA expression. Error bars = SEM. Analysis of variance with Tukey’s for post hoc analysis. Ns, not significant, ∗ = P < .05. While hypoxia is the canonical activator of HIFs, the models above utilize loss of VHL to drive accumulation and activation of HIF signaling rather than an actual hypoxia stimulus. Additionally, non-canonical pathways of HIF activation, especially in liver contexts, operate through mechanisms such as mitochondrial dysfunction and increased reactive oxygen species or succinate that inhibit upstream PHDs or inflammatory signaling such as nuclear factor-κB.37, 38, 39 To determine whether classical hypoxia can drive suppression of Hnf4α and the urea cycle, we incubated primary hepatocytes under normoxic and hypoxic conditions and evaluated expression of Hnf4α, Otc, and Cps1 by quantitative polymerase chain reaction. We utilized the AH2 mouse line, which carries a floxed Hif2α allele with a Cre-ERT2 recombinase under control of an albumin promoter. Primary hepatocytes were exposed to either 24 or 48 hours of 2% O2 or control normoxia; all time points were harvested 48 hours after isolation to control for dedifferentiation in culture. Treatment with hypoxia led to robust downregulation of Hnf4α at 24 and 48 hours of hypoxia, but not in HIF2-knockout hepatocytes. Additionally, expression levels of Otc and Cps1 were decreased at 48 hours of hypoxia, but not when HIF2 was deleted (Figure 5). This demonstrates that hypoxia in hepatocytes can drive suppression of Hnf4a and the rate-limiting urea cycle enzymes Cps1 and Otc and that this suppression is mediated by HIF2α. Hypoxia decreases expression of HNF4α, CPS1, and OTC in a HIF2α-dependent fashion. Primary hepatocytes from a floxed HIF2α line with albumin-driven Cre recombinase were isolated and treated with 2% O2 or normoxia for 24 or 48 hours, and gene expression was measured via quantitative polymerase chain reaction. All primary hepatocytes were isolated at the same total time from isolation to minimize differences from dedifferentiation in culture. N = 3 biological replicates. Analyzed via two-way analysis of variance with Sidak’s multiple comparisons test. NS, not significant, ∗∗ = P < .001, ∗∗∗ = P < .0001. Discussion In this study, we demonstrate that constitutive activation of HIF2α in hepatocytes leads to elevated ammonia levels in both the liver and serum, accompanied by suppression of urea cycle gene expression. Additionally, hypoxia in vitro is sufficient to suppress HNF4α, CPS1, and OTC in a HIF2α-dependent manner. Mechanistically, HIF2α downregulates the master hepatocyte transcription factor HNF4α, which regulates expression of key urea cycle enzymes, including OTC and CPS1. Importantly, these effects are specific to HIF2α, as deletion of HIF2α in the context of VHL loss normalizes ammonia levels, urea cycle enzyme expression, and HNF4α abundance. Collectively, these findings identify a previously unappreciated HIF2α-HNF4α axis that regulates nitrogen metabolism in the liver and provides a mechanistic link between hypoxia signaling and hyperammonemia in CLD. Prior work from our lab and others has shown that chronic HIF2α activation in hepatocytes promotes steatohepatitis and fibrosis,10,40, 41, 42 although the mechanisms driving this progression remain incompletely understood. Here, we show that an essential hepatocyte function, the urea cycle, is robustly downregulated in a HIF2α-dependent manner. OTC, an important regulatory enzyme of the cycle, is largely controlled by HNF4α.28 HIF2α-mediated suppression of HNF4α leads to increased serum and liver ammonia, which likely contributes to disease progression. Beyond its well-established role in HE in patients with decompensated cirrhosis, local elevations in ammonia may directly activate hepatic stellate cells, promoting inflammatory and fibrotic programs.17,19 This provides a plausible mechanism linking HIF2α activation to the accelerated fibrosis observed in our chronic hypoxia model. Our findings also raise the possibility that HIF2α promotes hepatocyte dedifferentiation. During acute injury or in the context of the acute-phase response, hepatocytes can revert to a liver progenitor cell-like state to reduce metabolic load and preserve survival.43 In chronic injury, however, this dedifferentiation is often sustained, resulting in loss of terminal hepatocyte function, ongoing inflammation, and fibrosis.44 HIF2α may exacerbate this process by suppressing hepatocyte-specific transcription factors such as HNF4α, thereby impairing the terminal differentiation program and perpetuating a maladaptive, pro-fibrotic state. Future studies will be important to define the specific pathways by which HIF2α promotes hepatocyte dedifferentiation, the extent of disruption in terminal differentiation, and whether restoration of key transcription factors in vivo can mitigate these effects. Beyond nitrogen metabolism, HIF2α activation perturbs lipid handling and other metabolic programs, suggesting a coordinated reprogramming of hepatocyte function during chronic liver injury.45,46 The interplay between impaired ammonia detoxification and dysregulated lipid metabolism may create a feed-forward loop that amplifies hepatocyte stress, inflammation, and fibrosis. Understanding how HIF2α orchestrates these metabolic and differentiation programs may therefore reveal multiple avenues for intervention in CLD. Conclusion In summary, our study identifies HIF2α as a central regulator of ammonia metabolism through suppression of HNF4α and downstream urea cycle enzymes. HIF2α-mediated alterations in nitrogen handling and hepatocyte differentiation likely contribute to the pathogenesis of steatohepatitis and fibrosis. These findings highlight the HIF2α-HNF4α axis as a potential therapeutic target in MASLD and CLD and suggest that modulation of this pathway could provide new biomarkers for risk stratification in patients prone to progressive liver disease. Acknowledgments Authors’ Contributions Marc T. Cantwell and Yatrik M. Shah conceived and designed the study. Marc T. Cantwell and Marwa O. El-Derany developed the methodologies. Marc T. Cantwell, Marwa O. El-Derany, Raja Gopal Reddy Mooli, and Sadeesh K. Ramakrishnan acquired the data. Marc T. Cantwell and Yatrik M. Shah analyzed and interpreted the data. Marc T. Cantwell and Yatrik M. Shah supervised the study and wrote the manuscript. All authors edited and provided input to the manuscript. Footnotes Conflicts of Interest: The authors disclose no conflicts. Funding: This work was supported by NIH grants R01CA148828, R01CA245546, and R01DK095201 to Yatrik M. Shah; the Center for Gastrointestinal Research DK034933; the Department of DefenseCA171086 to Yatrik M. Shah. Sadeesh K. Ramakrishnan was supported by NIH grants DK133406 and DK134581. Marc T. Cantwell was supported by NIH grant T32DK094775. Ethical Statement: The study did not require the approval of an institutional review board. Data Transparency Statement: All raw data supporting the findings of this study are available from the corresponding author upon request. Mouse lines utilized in this study are available upon request and subject to a material transfer agreement (MTA). Reporting Guidelines: ARRIVE guidelines 2.0 for animal studies. Material associated with this article can be found, in the online version, at https://doi.org/10.1016/j.gastha.2026.101039. Supplementary Materials Supplementary materials Extended PDF References Associated Data Supplementary Materials Supplementary materials Extended PDF