Cinnabarinic acid protects against metabolic dysfunction-associated steatohepatitis by activating Aryl hydrocarbon Receptor-dependent AMPK signaling

INTRODUCTION

Metabolic dysfunction-associated steatotic liver disease (MASLD) is a chronic hepatic disease with a spectrum of pathological conditions ranging from hepatic steatosis to metabolic dysfunction-associated steatohepatitis (MASH) with advanced fibrosis (cirrhosis) which can ultimately progress to hepatocellular carcinoma (HCC) (1). MASH is a debilitating hepatic disease associated with steatosis, hepatocyte apoptosis, inflammation, and fibrosis – and is highly prevalent among obese and type 2 diabetic patients (2). Except for a recent accelerated approval of thyroid hormone receptor beta agonist, resmetirom for patients with moderate to advanced fibrosis, there are no other FDA-approved pharmacological interventions (3). Therefore, identifying novel therapeutic treatments against the pathogenic components of MASH, particularly steatosis and fibrosis remains an intense focus of investigation.

Cinnabarinic acid (CA), a byproduct of a kynurenine pathway of tryptophan metabolism, is derived through the condensation of two molecules of 3-hydroxyanthranilic acid (3HAA) (4). Conversion of 3HAA to CA is catalyzed by enzyme cinnabarinate synthase in the liver (5). However, there is also an evidence for the non-enzymatic oxidation of 3HAA to CA (6). Previous studies have identified CA as an endogenous Aryl hydrocarbon Receptor (AhR) agonist (4, 7). AhR, a ligand-activated transcription factor, is historically studied within the context of xenobiotic metabolism (8–13). In a canonical AhR signaling, upon ligand interaction, AhR translocate to the nucleus, dimerizes with Aryl hydrocarbon Receptor nuclear translocator (Arnt), and interacts with the 5’-GCGTG-3’ motif termed xenobiotic response elements (XREs) present within the promoter of AhR responsive genes including those involved in phase I (Cyp1a1, Cyp1a2, Cyp1b1 etc.) and II (Gsta2, Ugt1a1, Ugt1a6 etc.) xenobiotic metabolism (10, 14–19). However, multiple previous studies have designated CA as a non-archetypical AhR agonist, as it does not induce expression of a prototypical AhR target gene, cytochrome P450, family 1, subfamily a, polypeptide 1 (Cyp1a1) in the liver but upregulates the expression of non-canonical AhR target genes (Stc2 etc.) (7, 20–22). Our laboratory has previously identified the hepatoprotective function of CA against a plethora of cellular and metabolic stressors. In isolated mouse primary hepatocytes, CA treatment protected against hepatocyte apoptosis induced by hydrogen peroxide, thapsigargin, and ethanol (21). Moreover, CA protected against both acute and chronic in vivo models of alcoholic liver disease in an AhR-dependent manner (21, 23). In monounsaturated free fatty acid-treated HepG2 and AML12 cells, which mimic hallmarks of MASLD pathology in vitro – CA attenuated steatosis and lipotoxicity (24). Additionally, in an in vivo diet-induced obesity model, preventative as well as therapeutic CA treatment was able to confer hepatoprotection (24, 25). However, CA’s potential to attenuate hepatosteatosis and fibrosis in an experimental model of metabolic dysfunction-associated steatohepatitis (MASH) is undetermined. Additionally, the involvement of hepatocyte AhR signaling and subsequent mechanism of CA-mediated protection is yet to be unraveled.

In the present study, to investigate the functional significance of hepatic AhR in CA-mediated protection, hepatocyte-specific AhR knockout (AhR-hKO) and AHR-floxed (AhR^fl/fl, control) mice were subjected to high-fat, high-fructose, high-cholesterol diet (MASH diet, MD) and treated with prophylactic CA regimen. Results revealed that AhR-hKO mice were more susceptible to MASH diet which resulted in increased hepatic steatosis and metabolic dysfunction compared to AhR-floxed mice. In AhR-floxed mice, CA treatment protected against hepatic steatosis, inflammation, hepatic injury, and fibrosis. Whereas, in the absence of AhR in hepatocytes, CA failed to confer protection against lipotoxicity - strongly suggesting the involvement of hepatic AhR signaling in CA-driven protection against MASH. Finally, our results demonstrate that the CA-induced AhR-dependent protection was mediated by the activation of hepatic AMPK signaling.

MATERIALS AND METHODS

RESULTS

DISCUSSION

AhR is a cytosolic, ligand-activated transcription factor belonging to the basic- helix-loop-helix Per-Arnt-Sim family (9, 11, 44–46). Since its discovery in the 1980s, AhR has been a major focus in toxicology due to its ability to mediate adaptive and toxic responses to a plethora of environmental contaminants including halogenated and polycyclic aromatic hydrocarbons (13, 47). While initial studies focused their attention on investigating AhR-dependent toxico-pathological pathways using a prototypical AhR agonist, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), multiple recent publications have indicated involvement of AhR agonists in altering the fate of metabolic diseases (14, 25, 47). Among them, CA – a catabolite of the kynurenine pathway of the tryptophan metabolism was identified as a non-prototypical AhR ligand (4, 48). We have previously reported CA’s ability to attenuate hepatic steatosis (24). However, given that CA is an endogenous AhR ligand, this study aims to determine the role of AhR in CA-mediated protection against steatohepatitis with fibrosis using experimental models of metabolic dysfunction-associated steatohepatitis (MASH). Herein, we demonstrate that CA-mediated protection is dependent on AhR signaling in hepatocytes. Moreover, CA-induced AhR upregulates AMPK signaling to induce PGC1α and decrease expression of the master regulator of lipid homeostasis, SREBP1 to mitigate lipogenesis and confer protection steatohepatitis.

This study revealed that hepatic deletion of AhR exacerbates the hallmarks of MASH under the western-style high-fat high-fructose high-cholesterol diet conditions. We observed that absence of AhR in hepatocytes resulted in aggravated steatosis, inflammation, and injury (Fig. 2). Moreover, histopathological and biochemical studies performed on AhR-hKO livers showed significant increase in periportal/perisinusoidal fibrosis (Stage 1/2) upon 20-week MASH-diet feeding (Fig. 2). A future study with an extended MASH-diet regimen (~ 30 weeks) will be performed to investigate significance of hepatic AhR in MASH with advanced fibrosis. Our results corroborated well with the previous observations where male hepatocyte-specific AhR knockout mice (AhR-hKO), when subjected to a high-fat diet (57.5 kcal%) exhibited increased hepatic steatosis and inflammation (49). Moreover, similar to previous studies that utilized high-fat-diet-fed male AhR-hKO and male tamoxifen-inducible hepatocyte-targeted AhR conditional knockout mouse model (AhR-iCKO), we did not observe significant changes in body weight gain or adiposity in MASH-diet-fed Ahr-hKO mice compared to AhR-floxed mice (49, 50). Previously, female AhR-iCKO mice when fed with a 60 kcal% high-fat diet for 15 weeks have shown significant reduction in body mass gain with multilocular and smaller white fat cells compared to female AhR-floxed mice (49, 50). Therefore, future studies in females are warranted where ablation of hepatocyte AhR in adult mice using tamoxifen-inducible Cre or AAV8- thyroxine binding globulin (TBG)-Cre will be utilized to untangle the metabolic complexities of AhR signaling in sexual dimorphism.

One of the most critical findings from this study was the observation that CA treatment protected against MASH only in the presence of hepatocyte AhR. Our histopathological observations indicated amelioration of steatosis, inflammation, fibrosis, and liver injury only in AhR-floxed but not in AhR-hKO mice (Fig. 2). Additionally, CA treatment resulted in decreased body weight gain, adiposity, and increased glucose tolerance in AhR-floxed mice (Fig. 1–3). A previous study has shown CA significantly reduced fat deposition in white adipose tissues and livers, as well as increased brown adipose tissue thermogenic protein expression without reducing food intake in a circadian rhythm disorder induced obesity model (51). CA specifically upregulated expression of hormone-sensitive lipase, adipose triglyceride lipase, peroxisome proliferator-activated receptor α, creatine kinase B, uncoupling protein 1 genes involved in thermogenesis and adipose tissue lipogenesis. Additionally, CA altered gut microbes and improved production of short-chain fatty acids in the circadian rhythm disorder induced obesity model (51). In a diet-induced obesity model, both preventative as well as therapeutic CA regimen decreased body mass gain (24). It is plausible that CA-induced AhR-dependent upregulation of a hepatokine, STC2 may play a role in adipogenesis. Accordingly, systemic treatment of recombinant STC2 protein to C57BL6 mice and leptin-deficient (ob/ob) mice resulted in body weight reduction, suggesting the reduction in adiposity may be locally induced in gut or adipose tissues rather than a centrally mediated mechanism on feeding behavior. STC2 administration decreased mRNA expression of adipogenic genes peroxisome proliferator-activated receptor ϒ, adipocyte protein 2, CEBPA CCAAT-enhancer binding protein α, and CEBPA CCAAT-enhancer binding protein β (52). However, the precise signaling mechanism of CA-mediated anti-obesity needs to be further determined. AhR has been known to have both beneficial and detrimental function in fatty liver disease depending on the selection of the ligands (25, 53). High-fat diet fed C57BL6 mice, when treated with AhR agonist TCDD, significantly upregulated stearoyl coenzyme decarboxylase 1 (Scd1), elevated de novo lipogenesis, increased hepatic triacylglycerol content and resulted in severe steatosis and fibrosis (54). On the contrary, AhR activation by gut-microbiota-derived tryptophan metabolites alleviated lipogenesis and mitigated MASLD (55–58). Indole-3-acetate administration downregulated fatty acid synthesis and expression of sterol regulatory element-binding protein 1 (Srebp1) in hepatocytes in an AhR-dependent manner (55, 58). Similarly, indole-activated AhR signaling pathway alleviated hepatic steatosis (57). Here, we showed that administration of AhR agonist CA, downregulated expression of acetyl-CoA carboxylase 1 (Acc1), fatty acid synthase (Fasn1), glycerol-3-phosphate acyltransferase 1 mitochondrial (Gpat1), glycerol-3-phosphate acyltransferase 2 (Gpat2), diacyl glycerol acyl transferase 2 (Dgat1), monoacylglycerol O-acyltransferase 1 (Mogat1), Scd1, and Srebp1 genes involved in fatty acid and triacylglycerol metabolism, decreased hepatic cholesterol, triacylglycerol, fatty acid content, and mitigated steatosis in livers of AhR-floxed mice (Fig. 2 and 5). However, CA failed to protect against lipotoxicity in vivo in AhR-hKO mice and in vitro in AhR-silenced human primary hepatocytes strongly indicating that hepatocyte AhR is obligatory for exerting CA’s anti-steatotic function (Fig. 2, 5-6).

Mechanistically, we identified CA-induced AhR-dependent AMPK phosphorylation in both mouse liver tissues and human hepatocytes (Fig. 5, 6). Multiple previous studies using genetic and pharmacological activators of AMPK have shown beneficial effects against diet-induced obesity and MASLD (59–62). However, role of AhR and its agonists in AMPK activation has not been extensively studied. A prior report using a liver-specific, constitutively activated human AhR transgenic mouse model showed upregulation of AMPK in liver and muscles. These mice were protected from high-fat-diet-induced obesity and showed insulin resistance but did not show attenuation of liver steatosis (63). Similarly, induction of AhR by kynurenine, a tryptophan catabolite upstream of cinnabarinic acid, upregulated protein phosphatase 2 regulatory subunit-Bdelta (PPP2R2D), decreased AMPK phosphorylation, autophagy and resulted in disturbed lipid metabolism in liver (64). Contrarily, treatment with a bona fide AhR agonist, kynurenic acid – which like CA, belongs to the kynurenine pathway of tryptophan catabolism – attenuated MASLD by the activation of AMPK-oxygen-regulated protein 150 pathway and the suppression of Srebp1 (65). However, we did not observe alterations in PPP2R2D and oxygen-regulated protein 150 expression upon CA treatment (data not shown). In non-hepatic tissues, selective AhR ligands have shown to upregulate AMPK, attenuate transcriptionally active mSREBP1, and repress lipid synthesis (66). These paradoxical effects are potentially due in part to the specificity of the agonists, degree of AhR expression and activation, recruitment of specific co-activators/co-repressors by AhR at chromatin complex, crosstalk of AhR with other transcription factors – resulting in the regulation of specific downstream signaling pathways involved in lipid metabolism (25, 67). In the context of these reports, we scrutinized the effect of CA-activated AhR on AMPK phosphorylation and demonstrated that treatment with CA increased AMPK phosphorylation in an AhR-dependent manner and pharmacological suppression of AMPK phosphorylation with compound C lost CA’s ability to mitigate steatosis (Fig. 5). These data strongly suggest that CA alleviates lipid accumulation in AhR-expressing hepatocytes at least in part via an AMPK-dependent pathway. However, given a presence of xenobiotic response elements on human and mouse genes encoding α, β, and ϒ subunits of AMPK, it is pertinent to investigate whether AMPK activation by CA-induced AhR is due in part by direct transcription regulation or indirect signaling mechanism in future studies.

As a master regulator of metabolism, AMPK is known to directly phosphorylate and activate PGC1α to modulate mitochondrial biogenesis (68). Overexpression of PGC1α is involved in the anti-inflammatory effects of AMPK through the downregulation of inflammatory proteins and cytokines including tumor necrosis factor alpha (TNFα), nuclear factor of kappa light polypeptide gene enhancer in B cells 1 (NFκB), Cyclooxygenase-2 (COX2), NLR family, pyrin domain containing 3 (NLRP3), interleukin 1 beta (IL1β), interleukin 6 (IL6) (Fig. 2, 4) (69). AMPK also inhibits de novo lipogenesis by inhibiting Acc1 that decreases malonyl-CoA, which is an allosteric inhibitor of Cpt1 (70). We have previously shown induction of Cpt1 with CA treatment (24), and here our results indicate CA-driven AhR-dependent upregulation of PGC1α phosphorylation (Fig. 5). AMPK has also shown to directly phosphorylate SREBP1 which inhibits its proteolytic cleavage from the precursor to mature form and thus dampen its ability to upregulate Acc and Fasn to increase de novo lipogenesis (71). Our data has shown significant downregulation of Srebp1 and its targets Acc, and Fasn with CA treatment in AhR-floxed livers (Fig. 5). Taken together, AhR-dependent activation of AMPK signaling in hepatocytes plays a crucial role in driving CA-mediated protection against steatosis.

In summary, our data suggests that hepatocyte-specific ablation of AhR results in aggravated hallmarks of steatohepatitis and fibrosis in a diet-induced model of MASH. Moreover, an endogenous AhR agonist, CA upregulates AhR-dependent AMPK signaling to protect against steatosis both in vitro human and in vivo mouse models of MASLD. Finally, further pre-clinical studies are warranted to investigate whether CA may serve as a lead compound with enhanced anti-MASH efficacy and reduced treatment-associated adverse toxicity in MASH models.

Supplementary Material

Supplemental material can be found here: doi.org/10.6084/m9.figshare.28439975

DATA AVAILABILITY

RNA-seq data will be made publicly available via NCBI Gene Expression Omnibus (GEO) at the time of publication of the manuscript (GEO accession number GSE277498). Study materials will be made available to other researchers from the corresponding author on reasonable request.