Pharmacological blockage of the AHR-CYP1A1 axis: a call for in vivo evidence.

AHR is a prognostic marker for aggressive cancer progression. High AHR levels have been found in many solid cancer types, including glioblastoma, ovarian cancer, and lung cancer. AHR might also have a direct effect on cancer at various stages: cell proliferation, tissue invasion, angiogenesis, inflammation, and metastasis. Blocking AHR is a promising approach to cancer immunotherapy (reviewed in [ 163 – 165 ]). BAY2416964 represents an orally active antagonist of AHR with IC 50 of 341 nM. An ongoing clinical trial with this small molecule is being conducted in patients with advanced cancer. By blocking AHR, it is expected that the oral administration of this small molecule will activate immune response against the tumor cells. This open-label, phase 1, first-in-human, dose-escalation, and dose-expansion study will evaluate the safety, tolerability, maximum tolerated or administered dose, pharmacokinetics, pharmacodynamics, and tumor response profile in patients with non-small cell lung cancer (NSCLC), head and neck squamous cell carcinoma (HNSCC), and colorectal cancer microsatellite stable (MSS). The compound was extracted from patent WO2018146010A1, and the study started in August 2019 and is at the recruiting phase [ 166 ]. A phase 1, open-label, dose-escalation, and dose-expansion study of IK-175 is being conducted in patients with locally advanced or metastatic solid tumors and urothelial carcinoma. This oral antagonist of AHR will be investigated in adult subjects diagnosed with any form of an advanced or metastatic solid tumor especially in patients who do not fully benefit from standard-of-care, including the checkpoint inhibitors. Safety and tolerability of IK-175, to determine the recommended phase 2 dose, will be assessed in addition to pharmacokinetics, pharmacodynamics, and biomarkers of response [ 167 ]. Ezutromid is a clinical stage compound for Duchenne muscular dystrophy patients. This compound was developed aiming at an increased expression of utrophin to mimic the missing dystrophin in this condition [ 168 ]. Ezutromid is an orally administered antagonist of AHR [ 169 ]. The drug undergoes extensive first-pass metabolism leading to low oral bioavailability. Drug absorption increased with lipid-rich diet when compared with fasted conditions and the absorption profile manifested secondary peaks in some patients. Ezutromid followed a biphasic elimination [ 170 ]. Ezutromid was discontinued in 2018 after failing to show any benefit as a modifying disease drug in phase II trial [ 171 ].

This review highlights the paucity of in vivo data on the effects of AHR antagonists and CYP1A1 inhibitors in rodent models of disease, namely the compound´s pharmacodynamics and tissue specificities for the mechanism of action; dose–effect dependence, AHR target genes, long term, and toxicity use. The experimental design of these studies relied mostly on prevention protocols. Most studies administered the compound orally, but pharmacokinetic properties and dose–response relationship were barely evaluated. The age of animals varied from infant to young adult ages (3 to 12 weeks for mice and 6 to 16 weeks for rats), and there was no study in elderly. Sex differences were also underrepresented, as only 40% of the studies with the AHR antagonists and 10% of studies with CYP1A1 inhibitors were performed in female animals. The mean time of drug administration was 3.6 ± 0.73 and 3.2 ± 0.44 weeks for mice and rats, respectively. The longer period of administration was investigated for α-NF, an AHR antagonist, that was used for 26 weeks in a mice model of obesity (Tables 1 and 2 ).

Over the past few years, a multiplicity of important functions of AHR has been found, surpassing its original role as a xenobiotic sensor and regulator of xenobiotic detoxification. In fact, AHR has been confirmed as an important signaling molecule regulating and maintaining homeostasis in different cells, tissues, and organs. Novel data highlights AHR-CYP1A1 axis activation in the mechanisms of disease, justifying the putative value of its therapeutic blockade. Thus, it is timely to investigate and better characterize the pharmacological properties of blockers of the AHR-CYP1A1 axis. AHR-CYP1A1 blockers might be useful in the treatment of chronic diseases including diabetes, hypertension, skeletal muscle disorders, and cancer, diseases known for being poorly controlled with the currently available drugs. For that, compiling evidence of pre-clinical pilot studies performed so far with AHR-CYP1A1 blockers is needed. Notably, many drugs identified in vitro demonstrated to be auto-inducers of its own metabolism in vivo, and this effect was tissue/organ specific. Most of the tested compounds ameliorated the disease although some potential adverse reactions might be also anticipated. Summing up, we believe the information herein compiled will be helpful to guide researchers when planning experiments with AHR-CYP1A1 blockers. Finally, and in view of the ongoing clinical studies using three AHR antagonists, we reinforce a call for further evidence on the pharmacological properties of AHR-CYP1A1 blockers.

The aryl hydrocarbon receptor (AHR) is a ligand-activated transcription factor that belongs to the basic helix-loop-helix/Per-Arnt-Sim (bHLH/PAS) family with important functions in sensing and incorporating environmental and outer stimuli (light–dark, O 2 alterations, xenobiotic exposure, and microbiota metabolites) into cellular adaptive responses (reviewed by [ 1 ]). Other members of this family are the CLOCK-BMAL1 (key components of the circadian clock), the hypoxia-inducible factors (HIFs), and the aryl hydrocarbon receptor translocator (ARNT, also named HIF-1β), and interesting evidence on the cross-play among them have been reported (reviewed by [ 2 ]). In the present review, a brief introduction will be given to the AHR-CYP1A1 axis, but we would like to invite the reader to find more information about these druggable targets in the excellent reviews suggested throughout this introductory note. As a cytosolic protein, the AHR exists in an inactive state bound to the chaperones heat shock protein 90 (HSP90), HSP90-associated co-chaperone p23, AHR-interacting protein (AIP), also called hepatitis B virus X-associated protein 2 (XAP2), and tyrosine kinase c-Src. This chaperone complex maintains the proper folding and assures the ligand-binding competency and, overall, the transcriptional effectiveness of AHR (reviewed by [ 3 , 4 ]). AHR ligands come from the environment and from the microbiota and cellular metabolism. There is a wide variety of compounds with different chemical properties, structure, and binding affinities, which have been recognized as AHR ligands. Among the exogenous compounds, it may be referred environmental contaminants including the halogenated aromatic hydrocarbons (HAH), such as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), 2,3,7,8-tetrachlorodibenzofuran (TCDF), and 3,4,3′,4,′5-pentachlorobiphenyl (PCB), and the polycyclic aromatic hydrocarbons (PAH), such as benzo[a]pyrene (B[a]P) and 3-methylcholanthrene (3-MC). In addition, some dietary substances have been also described including polyphenols (quercetin, resveratrol, curcumin, indole-3-carbinol) (reviewed by [ 5 , 6 ]). Among the endogenous ligands, some ultraviolet photoproducts of tryptophan have been described as the 6-formylindolo[3,2-b]carbazole (FICZ); the indigoids indigo and indirubin; the kynurenine (Kyn) and its metabolites including kynurenic acid; the metabolites of arachidonic acid like lipoxin 4A, prostaglandin G2, and hydroxyeicosatetraenoic acid; and cystine [ 7 ] and the tetrapyrroles derived from heme, biliverdin, and bilirubin (reviewed by [ 8 , 9 ]). This vast ligand promiscuity converges in the fact that according to the ligand, AHR might be activated in a diverse manner and in a cell- and tissue-dependent way [ 3 , 5 ]. AHR has long been associated with an adaptive response to environmental contaminants, most of which are man-made and not related to human physiology. This response activates the AHR canonical or adaptive pathway, which involves AHR-ARNT heterodimer binding DNA at xenobiotic response elements (XRE), including xenobiotic enzymes and transporters that allow environmental contaminants detoxification (reviewed by [ 4 , 10 , 11 ]). The CYP1A1 expression is a sensitive marker of the activation of this route [ 12 ]. Moreover, a negative regulatory feedback mechanism is present, with the induction of AHR repressor (AHRR) gene, capable of competing with AHR for ARNT, originating a transcriptionally inactive heterodimer and thus repressing AHR transcriptional activity [ 13 ]. Alternatively, several exogenous ligands (e.g., polyphenols as resveratrol or quercetin) activate the AHR alternative pathway, triggering the expression of the antioxidant and anti-inflammatory paraoxonase 1 (PON-1) [ 11 ]. Besides ARNT, the AHR also has non-canonical dimerization partners that will activate the transcription of non-consensus xenobiotic response elements (NC-XRE). One of these novel proteins is the tumor suppressor Kruppel-like factor 6 (KLF6) that is involved in the regulation of cellular differentiation, proliferation, and apoptosis [ 14 ]. Also, the AHR can interplay with other proteins, most notably with NF-kB subunits and ERα. At final steps, the AHR undergoes proteasomal degradation (Fig.  1 ). Fig. 1 Representations of ligand-dependent AHR activation pathways. Inactive AHR is localized at the cytoplasm complexed to some chaperones and other elements (HSP90, XAP2, p23, c-Src). Upon ligand binding, conformational changes allow AHR to translocate to the nucleus, where it dissociates from its chaperone complex. Depending on the ligand, AHR can follow either ARNT-dependent or ARNT-independent pathways. In the ARNT-dependent pathways, the AHR-ligand complex dimerizes with its binding partner ARNT. The ligand-AHR-ARNT complex will activate or repress the expression of different genes depending on the type of ligand. Canonical pathway (left side) . AHR binding to Kyn, HAH (e.g., dioxin), or PAH (e.g., B[a]P) activates the canonical pathway; the dimer AHR-ARNT binds to xenobiotic response elements (XRE) and drives the expression of xenobiotic metabolizing enzymes such as CYP1A1, the prototypical target gene of AHR. The AHR-ARNT complex promotes gene expression by recruiting several components of the transcriptional machinery and fulfilled this function; AHR activity ends by the dissociation of the complex from the DRE to be exported from the nucleus, where it enrolls in ubiquitin-mediated proteasomal degradation [ 26 ]. Moreover, a negative regulatory feedback mechanism is present, with the induction of AHR repressor (AHRR) gene, capable of competing with AHR for ARNT, originating a transcriptionally inactive heterodimer, thus repressing AHR transcriptional activity [ 13 ]. Alternative pathway (center) . AHR binding to polyphenols such as resveratrol or quercetin activates the alternative pathway by binding to alternative xenobiotic response elements such as the antioxidant and anti-inflammatory paraoxonase 1 (PON-1). Other non-canonical pathways (right side) include alternative REs of alternative AHR-ARNT related transcriptional responses also observed in tyrosine hydroxylase [ 27 ], a precursor enzyme in the synthesis of dopamine and catecholamines, Bax, an apoptosis regulator gene [ 28 ], and in TGF-β [ 29 ]. Also, the AHR can bind to other non-ARNT binding partners, like the Kruppel-like factor 6 (KLF6) or the NF-kB subunits RelA or RelB. ALHD3, aldehyde dehydrogenase 3; AHR, aryl hydrocarbon receptor; AHRE, aryl hydrocarbon response element; AHRR, AHR repressor; ARNT, aryl hydrocarbon receptor nuclear translocator; B[a]P, Benzo[a]pyrene; CYP1A1, cytochrome P450, family 1, subfamily A, polypeptide 1; CYP1A2, cytochrome P450, family 1, subfamily A, polypeptide 2; CYP1B1, cytochrome P450, family 1, subfamily B, polypeptide 1; GST, glutathione S -transferase; HAH, halogenated aromatic hydrocarbon; HSP90, heat shock protein 90; KLF6, Kruppel-like factor 6; Kyn, kynurenine; NC-XRE, non-consensus xenobiotic responsive element; p21cip1, cyclin-dependent kinase inhibitor 1; p23, HSP90-associated co-chaperone; PAI-1, plasminogen activator inhibitor 1; PAH, polycyclic aromatic hydrocarbon; PON1, paraoxonase 1; RelA, nuclear factor NF-kappa-B P65 subunit; RelB, RELB proto-oncogene, NF-kB subunit; TH, tyrosine hydroxylase; XAP2, hepatitis B virus X-associated protein 2 or aryl hydrocarbon receptor-interacting protein (AIP); XRE, xenobiotic responsive element Representations of ligand-dependent AHR activation pathways. Inactive AHR is localized at the cytoplasm complexed to some chaperones and other elements (HSP90, XAP2, p23, c-Src). Upon ligand binding, conformational changes allow AHR to translocate to the nucleus, where it dissociates from its chaperone complex. Depending on the ligand, AHR can follow either ARNT-dependent or ARNT-independent pathways. In the ARNT-dependent pathways, the AHR-ligand complex dimerizes with its binding partner ARNT. The ligand-AHR-ARNT complex will activate or repress the expression of different genes depending on the type of ligand. Canonical pathway (left side) . AHR binding to Kyn, HAH (e.g., dioxin), or PAH (e.g., B[a]P) activates the canonical pathway; the dimer AHR-ARNT binds to xenobiotic response elements (XRE) and drives the expression of xenobiotic metabolizing enzymes such as CYP1A1, the prototypical target gene of AHR. The AHR-ARNT complex promotes gene expression by recruiting several components of the transcriptional machinery and fulfilled this function; AHR activity ends by the dissociation of the complex from the DRE to be exported from the nucleus, where it enrolls in ubiquitin-mediated proteasomal degradation [ 26 ]. Moreover, a negative regulatory feedback mechanism is present, with the induction of AHR repressor (AHRR) gene, capable of competing with AHR for ARNT, originating a transcriptionally inactive heterodimer, thus repressing AHR transcriptional activity [ 13 ]. Alternative pathway (center) . AHR binding to polyphenols such as resveratrol or quercetin activates the alternative pathway by binding to alternative xenobiotic response elements such as the antioxidant and anti-inflammatory paraoxonase 1 (PON-1). Other non-canonical pathways (right side) include alternative REs of alternative AHR-ARNT related transcriptional responses also observed in tyrosine hydroxylase [ 27 ], a precursor enzyme in the synthesis of dopamine and catecholamines, Bax, an apoptosis regulator gene [ 28 ], and in TGF-β [ 29 ]. Also, the AHR can bind to other non-ARNT binding partners, like the Kruppel-like factor 6 (KLF6) or the NF-kB subunits RelA or RelB. ALHD3, aldehyde dehydrogenase 3; AHR, aryl hydrocarbon receptor; AHRE, aryl hydrocarbon response element; AHRR, AHR repressor; ARNT, aryl hydrocarbon receptor nuclear translocator; B[a]P, Benzo[a]pyrene; CYP1A1, cytochrome P450, family 1, subfamily A, polypeptide 1; CYP1A2, cytochrome P450, family 1, subfamily A, polypeptide 2; CYP1B1, cytochrome P450, family 1, subfamily B, polypeptide 1; GST, glutathione S -transferase; HAH, halogenated aromatic hydrocarbon; HSP90, heat shock protein 90; KLF6, Kruppel-like factor 6; Kyn, kynurenine; NC-XRE, non-consensus xenobiotic responsive element; p21cip1, cyclin-dependent kinase inhibitor 1; p23, HSP90-associated co-chaperone; PAI-1, plasminogen activator inhibitor 1; PAH, polycyclic aromatic hydrocarbon; PON1, paraoxonase 1; RelA, nuclear factor NF-kappa-B P65 subunit; RelB, RELB proto-oncogene, NF-kB subunit; TH, tyrosine hydroxylase; XAP2, hepatitis B virus X-associated protein 2 or aryl hydrocarbon receptor-interacting protein (AIP); XRE, xenobiotic responsive element To add more complexity, several lines of research have documented ligand-independent AHR activation, namely modulation by cAMP [ 15 ], oxidized LDL [ 16 ], vascular shear stress [ 17 ], or even reactive oxygen species (ROS) [ 18 ]. The mechanism by which AHR ligands affect physiological processes appears to involve multiple interactions between AHR and other signaling pathways (reviewed extensively by [ 3 , 4 , 10 , 11 ]) or may be a consequence of a change in the activity of metabolic enzymes and transporters, which are AHR target genes that modify the availability and disposition of endogenous metabolites (e.g., estrogen, arachidonic acid, melatonin) [ 19 – 21 ]. AHR target genes include several drug-metabolizing enzymes belonging to the cytochrome P450, like cytochrome P450, family 1, subfamily A, polypeptide 1 (CYP1A1), CYP1A2 and CYP1B1, aldehyde dehydrogenase 3 (ALDH3), and UDP glucuronosyltransferase family 1 member A1 (UGT1A1) [ 22 – 24 ] and transporters such as the ATP binding cassette subfamily B member 1 (ABCB1) [ 25 ].

The anthraquinone derivative alizarin (1,2-dihydroxyanthraquinone) is a food pigment described as a strong competitive inhibitor of CYP1A1 and CYP1A2 in vitro [ 126 ]. Alizarin also inhibits CYP1B1 and to a lesser extent CYP2A6 and CYP2E1 in vitro [ 126 ]. The antioxidant activity of alizarin was tested in a mice model of hepatotoxicity induced by bromobenzene. Pre-treatment with alizarin reduced the hepatic lipid peroxidation (thiobarbituric acid reactive substances (TBARS) assay) and serum aspartate aminotransferase (AST) levels [ 127 ]. CYP1A1 was not investigated. Takahashi and collaborators (2007) evaluated the preventive effect of alizarin on MeIQx (amine 2-Amino-3,8-dimethylimidazo[4,5-f]quinoxaline)-induced DNA adducts. C57BL/6 N male mice (5-weeks-old) were administered with the anthraquinone for 3 days followed by MeIQx alone or in combination with alizarin for 3 days. Pre-treatment with alizarin showed only a marginal reduction in the amount of adducts present in the lung and kidney. Moreover, the EROD assay (CYP1A activity) remained unchanged [ 128 ]. Although the association with CYP1A1 was not investigated, Xu and co-workers (2019) showed that alizarin significantly decreased the levels of blood glucose, ameliorated lipid metabolism abnormalities, and decreased oxidative stress in a diabetic mice model upon administration of alizarin to male Kunming mice for 10 days [ 129 ]. Ellipticine (5,11-dimethyl-6 H -pyrido[4,3- b ] carbazole) is a polyaromatic alkaloid that inhibits DNA topoisomerase II and forms covalent adducts with DNA. Ellipticine has been associated with anti-tumoral activity in vitro [ 130 ]. Ellipticine is metabolized mainly by CYP1A1/2 and CYP3A4 in vitro [ 131 ]. Although in vitro studies classify ellipticine as an inhibitor of CYP1A1, CYP2B, and CYP3A [ 132 ], in vivo evidence shifts to different conclusions. In a dose–response study, where male and female Wistar rats were treated with a single dose of 4, 40, or 80 mg/kg of ellipticine, Aimová and collaborators (2007) found sex differences in CYP1A1 content in the liver upon ellipticine administration, with males being more responsive to this compound. The lowest dose of ellipticine (4 mg/kg) originated a 26-fold increase in CYP1A1 induction in males, differently to the moderate fivefold induction observed in females. Also, while in males, CYP1A1 induction correlated positively with ellipticine dose, in females, the maximum induction of CYP1A1 was attained with the intermediate dose (40 mg/kg). CYP1A1 protein and mRNA levels and activity were assessed in the lung, liver, and kidney of male rats, showing a clear induction of CYP1A1 in these organs. The middle dose induced the highest increase of mRNA levels in the lung (24-fold), followed by the kidney (tenfold) and the liver (fivefold) [ 133 ]. The effects of ellipticine on CYP1A1 activity were also evaluated in a time-response study. Animals were given a single dose of 80 mg/kg of ellipticine and sacrificed at four time points (2–224 days). CYP1A1 protein levels achieved their peak at day 2, returned to basal levels at day 14, and remained low until the end of follow-up [ 133 ]. More studies are needed to evaluate the effects of single and multiple dose administration of ellipticine on CYP1A1 in vivo. It also remains to be understood the mechanism by which ellipticine might induce CYP1A1 despite being described as a CYP1A1 inhibitor in vitro. Ellipticine reduced the tumor size and demonstrated its activity against cell proliferation in a model of non-small cell lung cancer [ 134 ] (Table 2 ). Ellipticine effectively prevented inflammation in the endotoxic shock mouse model: there was a time-dependent reduction in the levels of TNF-α and IL-6 [ 135 ] (Table 2 ). Pterostilbene is a cell-permeable stilbenoid, analog of resveratrol, originally derived from Pterocarpus marsupium . This compound has antioxidant, anti-proliferative, anti-inflammatory, and hypoglycemic effects and is a very potent inhibitor of CYP1A1 in vitro [ 136 ]. Pterostilbene also inhibits CYP2C8 and UGT1A6 enzyme activities in vitro [ 137 ]. Pterostilbene prevented edema and inflammatory markers in a carrageenan-induced inflammation mice model [ 138 ]. Like alizarin, purpurin is an anthraquinone derivative and a food supplement inhibitor of CYP1A1 and CYP1A2 in vitro [ 126 ]. Purpurin also inhibits CYP1B1 and to a lesser extent CYP2A6 and CYP2E1 in vitro [ 126 ]. Using a mice model of hepatotoxicity induced by bromobenzene (see Alizarin and Table 2 ), purpurin showed a time-dependent decrease in the activity of CYP1A1 and the formation of MelQx-DNA adducts at the lungs, kidney, and marginally at the liver. Purpurin displayed a stronger inhibitory capacity for CYP1A1 than alizarin [ 128 ]. Although CYP1A1 activity was not evaluated, an anti-obesity effect has been described for this compound: a dose-dependent reduction in weight gain [ 139 ]. Rhapontigenin (3, 3′, 5-trihydroxy-4′-methoxy-stilbene) is a hydroxystilbene derivative, with similar structure to resveratrol, derived from the roots of Rheum undulatum [ 140 ]. This compound proved to be a mechanistic based inhibitor of the human CYP1A1 in vitro [ 141 ]. Rhapontigenin also inhibits CYP3A4 and CYP2C9 [ 142 ]. Pharmacokinetic investigation in rats showed that rhapontigenin has a half-life of 3 h and is extensively glucuronidated and predominantly cleared by the liver [ 143 , 144 ]. While no relation with AHR signaling was investigated, rhapontigenin was associated with cardioprotective effects in a model of isoproterenol-induced myocardial infarction in male Sprague–Dawley rats [ 145 ]. Rhapontigenin pretreatment ameliorated infarct size and heart weight and reduced protein expression of cardiac markers such as creatine kinase (CK), cardiac troponin-T (CTT), and lactate dehydrogenase (LD). It also reduced protein expression of superoxide dismutase (SOD) and malonaldehyde (MD), IL-6, p38, inducible NO synthase (iNOS), and TNF-α [ 145 ]. Rutaecarpine is a quinazolinocarboline alkaloid extracted from the fruit Evodia rutaecarpa and commonly found in herbal products [ 146 ]. Rutaecarpine is an example of a compound whose impact on CYP1A1 activation has been described with conflicting results in different studies. In vitro studies described the compound as a selective CYP1A1 inhibitor [ 147 ]. However, most of the in vivo evidence shows that rutaecarpine activates the AHR-CYP1A1 axis [ 148 – 151 ]. For instance, rutaecarpine increased EROD activity in the mice liver (sixfold) and hepatic CYP1A1 (western blot). This effect was not observed in the kidney, denoting organ differences in rutaecarpine modulation of CYP1A1 [ 148 ]. The effects of rutaecarpine in vitro and in vivo that support its putative cardiovascular protective effect are reviewed elsewhere [ 152 ]. Rutaecarpine was investigated in in vivo models of myocardial ischemia–reperfusion injury [ 153 ], hypoxia-induced right ventricular remodeling [ 154 ], arterial hypertension [ 155 – 158 ], atherosclerosis [ 159 ], arterial remodeling [ 160 ], obesity [ 161 ], and renal ischemia–reperfusion injury [ 162 ]. Despite the beneficial cardiovascular effects observed, none of these works linked rutaecarpine action to the AHR-CYP1A1 axis (Table 2 ).