1. Introduction

Underdevelopment of bilirubin-uridine diphosphate glucuronosyl transferase (UDPGT) is the major cause of neonatal hyperbilirubinaemia (NNH) [1]. Severe NNH can lead to kernicterus, a condition associated with neurological dysfunction and brain damage. Bilirubin exerts its toxicity by modulating immune and neuronal functions. Thus, bilirubin causes the release of the proinflammatory cytokines interleukins IL-1β and IL-6 and tumor-necrosis factor TNF-α [2]. These cytokines are powerful inducers of the extrahepatic tryptophan (Trp)-metabolising enzyme indoleamine 2,3-dioxygenase (IDO1) [3]. IDO1 induction leads to activation of the extrahepatic kynurenine (Kyn) pathway (KP) resulting in increased production of Kyn metabolites (Figure1), of which 3-hydroxykynurenine (3-HK), 3-hydroxyanthranilic acid (3-HAA) and quinolinic acid QA) undermine allogeneic T-cell proliferation and induce apoptosis, thus undermining T helper Th1 cells [4,5]. The Kyn transamination (KAT) product kynurenic acid (KA) is both pro- and anti-inflammatory [6] and is the endogenous antagonist of the N-methyl-D-aspartate (NMDA) receptors of the excitatory amino acid glutamate, in contrast with the agonist QA [7,8]. The balance between QA and KA determines the level of neuronal excitability: an important feature in kernicterus. Inhibition by bilirubin of aspartate aminotransferase [9], also known as KAT IV, raises the possibility that other KAT isoforms could also be inhibited, leading to decreased KA formation. This and the ability of IL-1β to activate kynurenine monooxygenase (KMO) and kynureninase [10] combine to drive the KP towards the above proinflammatory metabolites and shift the balance between QA and KA in favour of hyperglutamatergic activity and neuronal hyperexcitability. That acute bilirubin encephalopathy, a precursor of kernicterus, is associated with NMDA receptor activation has been reported [11]. Most brain QA is of peripheral origin and its circulating or cerebrospinal fluid levels are increased in a range of brain conditions associated with neuronal dysfunction and glutamatergic activation [12]. Bilirubin toxicity can however be modulated by the two products of the heme oxygenase 1 (HO1) reaction (Figure 1): biliverdin and carbon monoxide (CO), both of which exert protective effects: biliverdin antagonising oxidative damage and CO affording cell and tissue protection (see [13] and references cited therein). Bilirubin can also increase biliverdin levels by two mechanisms: (1) bilirubin oxidation by two cytochrome P-450 isoenzymes: P-450 2A5, which is induced by bilirubin, and P-450 2A6 [14]; (2) induction of HO1: an effect that can be reversed by bilirubin photoisomers [15]. Thus the balance between levels of bilirubin, biliverdin and CO is likely to determine extent of neuronal dysfunction in NNH.

Light therapy, the current gold standard NNH treatment, involves bilirubin photooxidation to less harmful or safe products. Bilirubin also possesses protective properties [9,16], but the fact that phototherapy detoxifies bilirubin suggests that the toxic effects outweigh the protective ones. The major product of bilirubin photooxidation is lumirubin [17], simple formation of which cannot explain the therapeutic outcome, as its levels are much lower than those of bilirubin. For example, whereas serum [bilirubin] in simple NNH is < 85 mM and could rise in severe cases to > 340 mM, serum lumirubin levels rise from 0.3 mM before, to only 2.5 mM after, phototherapy [18] and 11 hours after initiation of phototherapy, urinary lumirubin levels reach only 5.47 mM [19]. Lumirubin and possibly other photooxidation products must therefore exert powerful protective effects that can counteract the toxicity of considerably high levels of bilirubin. Lumirubin and other products exert no toxic effects on human neuroblastoma cells, nor influence cell cycle phase or expression of genes of bilirubin metabolism [20]. Very little is known of potential effects of lumirubin on immune markers, except its modest elevation of TNF-a protein expression at higher concentrations, compared to bilirubin, in unstimulated murine macrophage-like cells [21]. As will be discussed below, lumirubin acts as a ligand of receptors that can counteract the effects of bilirubin on immune and neuronal functions, making it a potential candidate underpinning the therapeutic activity of phototherapy. As the Trp metabolite 6-formylindolo[3,2-b]carbazole (FICZ), can also be formed by photooxidation of Trp [22,23], and given its protective effects in various disease models [24], it may play a protective role during NNH phototherapy. To explore the potential mechanisms of actions of lumirubin and FICZ in NNH, we examined in the present study using molecular docking in silico the likely binding of lumirubin, its 2 precursors biliverdin and bilirubin, and FICZ to a number of receptors involved in modulating immune and neuronal functions, namely the aryl hydrocarbon (AhR), N-methyl-D-aspartate (NMDA) and g-aminobutyric acid (GABA) receptors, the results of which provide pointers towards the therapeutic potentials of both lumirubin and FICZ in NNH and other conditions that merit exploration at the preclinical and therapeutic levels.

Figure 1: Diagrammatic representation of the interactions between bilirubin, tryptophan metabolism and the pathophysiology of NNH and kernicterus.

2. Methods

Molecular docking was performed using the Molegro virtual Docker (MVD) software as described previously [25,26]. Re-ranking was performed to improve accuracy. When re-ranked, 10 independent docking runs resulted in 10 solutions. The structures of lumirubin, biliverdin and bilirubin (Supplementary Figure S1) were imported into the MVD workspace in SDF format. All H atoms were added and important valence checked by using the utilities in MVD. The crystal structures of a number of human receptors were obtained from the protein data bank (http://www.rcsb.org/pdb) and imported into the MVD work region at reasonable resolution levels (≤ 2.8 Å). The following receptors were tested: AhR (PDB ID: 5Y7Y [27]); NMDA (PDB ID: 3JPW [28]); GABA_A (PDB ID: 4COF [29]); GABA_B (PDB ID: 4MS [30]); Kainate (PDB ID: 8FWR [31]). Docking was compared between bilirubin and lumirubin with 4 enzymes of the KP, namely Trp 2,3-dioxygenase (TDO) (PDB ID: 2NW8 [32]), IDO 1 (PDB ID: 5WHR [33]), Kyn monooxygenase (KMO) (PDB ID: 5X6Q [34]) and Kyn aminotransferase II (KAT II) (PDB ID: 5EUN [35]).

The ligands were prepared by assigning the missing bonds, hydrogen(s) and charges using UCSF Chimera software. Scoring functions of the MVD and other docking details have been described [26]. Docking of the above 3 compounds to the aryl hydrocarbon receptor (AhR) was compared with that of the known potent AhR ligands FICZ, indirubin and TCDD (2,3,7,8-Tetrachlorodibenzo-p-dioxin). Docking of lumirubin, its two precursors and FICZ was also performed with the NMDA, kainate, GABA_A and GABA_B receptors and compared with those of the well-established NMDA ligands KA and QA.

3. Results

3.1 Docking of biliverdin, bilirubin and lumirubin to the aryl hydrocarbon receptor (AhR)

The docking details of biliverdin, bilirubin and lumirubin to the crystal structure of the human AhR are given in Table 1 and illustrated in Figure 2. Amino acid residues at the AhR active site and those binding the ligands are listed in the legend to Figure 2. As shown in Table 1, neither biliverdin nor bilirubin docked to the AhR, whereas lumirubin docked very strongly (rerank score: -100.98 Kcal/mol). For comparative purposes, docking to the AhR of the currently recognised strong agonists indirubin (a constituent of Indigo naturalis and a product of bacterial metabolism of indoxyl sulphate), FICZ and TCDD is also included in Table 1. As shown, lumirubin docking was slightly stronger than those of these three powerful ligands.

Table 1: Diagrammatic representation of the interactions between bilirubin, tryptophan metabolism and the pathophysiology of NNH and kernicterus Abbreviations: AhR (aryl hydrocarbon receptor), CO (carbon monoxide), CBS (cystathionine β-synthase), FICZ (6-formylindolo[3,2-b]carbazole), GSH (reduced glutathione), HO1 (heme oxygenase 1), H2S (hydrogen sulphide), 3-HAA (3-hydroxyanthranilic acid), 3-HK (3-hydroxykynurenine), IDO (indoleamine 2,3-dioxygenase), IL (interleukin),  KA (kynurenic acid), KAT (kynurenine aminotransferase), KMO (kynurenine monooxygenase), QA (quinolinic acid), TNF (tumor-necrosis factor).

The heme released from hemoglobin breakdown is oxidised by heme oxygenase 1 to biliverdin plus CO.  The latter two compounds exert protective effects, but CO in excess could inhibit cystathionine β-synthase activity thereby inhibiting the formation of two protective compounds: reduced glutathione and hydrogen sulphide. Thus CO could exert positive and negative effects in kernicterus. Bilirubin can be back-oxidised to biliverdin by cytochromes P-450. The balance between bilirubin, biliverdin and CO may thus determine the extent of toxicity to the newborn. Bilirubin toxicity elicits a proinflammatory state that can impact the kynurenine pathway of tryptophan metabolism.  Cytokine induction of IDO1 results in production of kynurenine metabolites that undermine immune function and induce a hyperglutamatergic state, which can continue unabated through a vicious circle involving excessive activation of the AhR and IL-6, KA and IDO1. Phototherapy of NNH produces lumirubin, which could act by activation of the AhR, blockade of glutamate receptors and activation of GABA receptors.  The Trp photooxidation product FICZ could also be formed during NNH light therapy and may act in a manner similar to lumirubin. At small rapidly metabolised levels as would be expected during light therapy, FICZ can inhibit IDO1 induction by blocking the bilirubin elevation of proinflammatory cytokines.   

Figure 2: Docking of biliverdin, bilirubin and lumirubin to the human aryl hydrocarbon receptor (AhR). Amino acid residues at the AhR active site and those docked by ligands were as follows: biliverdin (Lys255, Arg267, Ser131, Ala132, Lys255, Gln254, Gly253, Leu251, Phe252, Pro266, Leu264) and (Lys256, Arg267) respectively; the corresponding residues with bilirubin docking were (Ser218, Val179, Ile312, Gln224, Lys313, Gly310, Phe177) and (Tyr311, Ser122) respectively; the corresponding residues with lumirubin were (Lys255, Gln254, Gly253, Leu251, Phe252, Pro266, Leu264) and (Gly253, Lys256) respectively.

3.2 Docking of biliverdin, bilirubin and lumirubin to the NMDA receptor

As summarised in Table 2, illustrated in supplementary Figure S2, and detailed in supplementary Table S1, neither biliverdin nor bilirubin docked to the NMDA receptor, whereas lumirubin docked strongly. Docking of the established NMDA ligands KA and QA was comparable, whereas that of FICZ was relatively stronger (Table 2).

Table 2: Rerank docking scores of biliverdin, bilirubin, lumirubin, FICZ, KA and QA to various receptors.

3.3 Docking of biliverdin, bilirubin and lumirubin to the GABA_A receptor

Neither biliverdin nor bilirubin docked to the GABA_A receptor, whereas both lumirubin and FICZ did (Table 2, Figure S3 and Table S2). Both KA and QA docked at comparable energy levels, slightly lower than those of lumirubin and FICZ.

3.4 Docking of biliverdin, bilirubin and lumirubin to the GABA_B receptor

As shown in Table 2, Figure S4, and Tables S3 and S4, similar docking levels were observed with FICZ, KA and QA, but slightly stronger with lumirubin.        

3.5 Docking of biliverdin, bilirubin and lumirubin to the kainate receptor

With the kainate receptor (Table 2, Figure S5 and Table S4), lumirubin showed the strongest docking followed by KA > QA, with FICZ being a weaker ligand. Both biliverdin and bilirubin failed to dock to the kainate receptor.

3.6 Docking of bilirubin and lumirubin to key enzymes of the kynurenine pathway

As shown in Table 3 and depicted in Figure 3, bilirubin failed to dock to any of the 4 KP enzymes Trp 2,3-dioxygenase (TDO), indoleamine 2,3-dioxygenase (IDO1), Kyn monooxygenase (KMO) and Kyn aminotransferase II (KAT II), whereas lumirubin docked strongly to all 4 enzymes.

Table 3: Re-rank docking scores of bilirubin and lumirubin to enzymes of the kynurenine pathway.

4. Discussion

Neonatal hyperbilirubinemia (NNH) is due to underdevelopment of bilirubin UDPGT [1], with phototherapy being the current standard treatment. The use of adjunctive phenobarbitone (an inducer of bilirubin UDPGT) makes little difference to the efficacy or speed of action of light therapy [36]. Whereas it is generally accepted that light therapy neutralises the toxicity of large concentrations of bilirubin and that the main photooxidation product lumirubin is not toxic, but can afford protection to the newborn infant, the precise mechanism(s) of its action is not fully understood. In the present study, we provide evidence from molecular docking in silico that lumirubin is one of the most powerful AhR ligands known, exhibiting a docking (binding) score greater than those of the potent ligands FICZ, indirubin and TCDD (Table 1). We also demonstrate the failure of bilirubin and biliverdin to dock to the crystal structure of the human AhR, thus providing the first test of direct interaction (or lack of it) of these 2 heme-degradation products with this receptor since 1998 and emphasising the need to assess binding of ligands to target proteins by the use of more direct pharmacological methods (see [16] and also the discussion on indoles [37]), rather than the widely used proxy system of P-450-related drug metabolism (see, e.g,[38]). Other alternative methods include the use of AhR^-/- KO models or molecular docking. Only the latter method tests ligand-protein interaction in the purest form free from the biological confounders that exist in cell cultures, tissue preparations or in vivo. As far as we could ascertain, ours is the first study to demonstrate by molecular docking the binding to the NMDA receptor of its main physiological modulators the antagonist kynurenic acid and agonist quinolinic acid and that both lumirubin and FICZ exhibit strong ligand binding to these receptors. We therefore suggest that a biological basis could be proposed for a role of lumirubin and possibly also FICZ in NNH phototherapy. In the following discussion focusing on immune and neuronal dysfunction in NNH, we examine the potential mechanisms of the proposed efficacy of these two light-induced compounds based on their known properties and their interactions with receptors.

4.1 Modulation of immune function

The aryl hydrocarbon receptor (AhR) is a transcription factor, activation of which can be beneficial, whereas its over activation is harmful [39]. It is described as “a sensor that modifies the immune response to the prevailing situation: reducing inflammation in an endogenous response situation, and, in the face of an exogenous threat, limiting damage by enhancing proinflammatory Th17 cell differentiation, antiinflammatory IL-22 generation and P-450-dependent metabolism of the foreign agent” [39]. For example, in experimental models of inflammation, the potent AhR agonist FICZ can either ameliorate or enhance severity. It can afford protection by decreasing levels of proinflammatory IL-17, IL-1β, IL-6, IFN-g and TNF-a and increasing those of IL-10 and IL-22, but it can also increase severity by promoting Th-17 cell differentiation (see [40] and references cited therein). These opposite effects of FICZ are determined by the route of administration and hence pharmacologically acting levels, with oral or parenteral administration being associated with decreased severity, presumably due to faster clearance and consequently low levels, and the subcutaneous route favouring slow and sustained absorption allowing continued activation of the AhR to induce a greater severity (see [24,40] and references cited therein). However, as phototherapy does not cause harm (see below), it is likely that during NNH phototherapy FICZ will not accumulate in large enough concentrations. The same applies to lumirubin especially in view of its rapid clearance [41] and albumin binding, however limited this is in infants. As is the case with FICZ, lumirubin can however exert toxic effects, but this has been demonstrated with a concentrations (25 mM) described by Jašprová et al. [42] as biologically-relevant, although the effects of lumirubin on proinflammatory cytokine gene expression were cell-dependent. Earlier, the same group [15] reported the inability of lumirubin at the above high concentration to influence the viability of SH-SY5Y cells, unlike bilirubin. As stated above, the rise in serum and urinary lumirubin after phototherapy is in the low mM range [18,19].

Although bilirubin does not interact with the AhR, it can activate it indirectly via proinflammatory Kyn metabolites after cytokine induction of IDO1. KA is the Kyn metabolite with the greatest affinity for the AhR [43]. One such cytokine whose levels are elevated by bilirubin, TNF-a, enhances gene expression of IL-6 [44]. IL-6 participates in the AhR control of IDO1 expression via the autocrine loop of AhR-IL-6-Stat3 signaling [45,46] and, although KA induces IL-6, the elevation of this cytokine by inflammation can induce IDO1 to produce sufficient amounts of KA to activate the AhR [43], thereby establishing a vicious circle. Bilirubin toxicity is associated with these changes 9]. The AhR also controls expression of poly (ADP-ribose) polymerase (PARP) that is involved in many important processes, including DNA repair [47,48]. A potential activation of NAD⁺-consuming PARP can result in cell death due to NAD⁺ and ATP depletion [49,50]. The preceding discussion suggests the need to assess the ability of lumirubin to counteract the above effects of bilirubin. Lumirubin however may act directly on enzymes of the KP, as our docking data (Table 3, Figure 3) suggest. It interacts with the 4 key KP enzymes TDO, IDO 1, KMO and KAT, but in which direction remains to be established in studies of enzyme activities and metabolite levels. Given its protective effect, lumirubin is likely to act as antagonist (inhibitor) of IDO1, TDO, KMO and possibly also KAT. Part of the lumirubin structure resembles that in a number of IDO1 inhibitors (see, e.g., [51]). If lumirubin is established as a dual TDO/IDO1 inhibitor, it may qualify as a potential cancer therapy, given the involvement of these 2 enzymes in cancer biology [52]. FICZ has already been shown in experimental cancer models to exhibit therapeutic potential [53-55].

Figure 3: Docking of bilirubin and lumirubin to kynurenine pathway enzymes. Amino acid residues at the enzyme active site and those binding the ligands were as follows: TDO: bilirubin (Thr254, Arg117, Ser123,124 and Gly125, Thr254, Ser124 respectively), lumirubin: (Leu129, 30, Gln130, Arg158, Ser128 and Ser128, Asp31) IDO 1: bilirubin (Arg296,297, Gln293, Arg303, Leu307 and Arg297, Gln293), lumirubin (Gln280, Ala283, Glu119, Gly284, Arg297 and Arg297, Thr 282). KMO: bilirubin (Pro402, Ser405, Arg386, Gln390, Tyr382, Arg403 and Arg100,84, Gly99), lumirubin (Arg111, Asn115, Gly16, Ala15, Arg39, Glu37, Phe13 and Phe131), KAT II: bilirubin (Ser17,77, Pro76, Gln289, Ile19, Pro18, Arg20 and Ala78, Ser17), lumirubin (Tyr233,142, Asn202, Ser117,260, Gln118 and Ser262, Tyr233, Arg270).

 4.2 Modulation of neuronal function

As the excitotoxicity in kernicterus involves glutamatergic activation, we have explored the possibility that docking to NMDA, kainate and GABA receptors by the above compounds may provide pointers to their toxic or protective effects. As neither bilirubin nor biliverdin interacts directly with the NMDA, kainate or GABA receptors, it is most likely that bilirubin-induced neuronal dysfunction is an indirect effect probably mediated by modulation of Trp metabolism by proinflammatory cytokines creating a hyperglutamatergic state that would be reflected in an increased [QA]/[KA] ratio and excitotoxicity. By contrast, lumirubin interacts directly with the above receptors, thereby providing a biological basis of its efficacy during NNH phototherapy against neuronal dysfunction. Whereas docking of KA and QA to the NMDA, GABA_A and GABA_B receptors was comparable, that of lumirubin was moderately greater (by 22.0%, 18.5% and 20.5% respectively). Given the opposite effects of KA and QA on NMDA receptors and the decreased toxicity of bilirubin upon photooxidation, it is not unreasonable to suggest that lumirubin may act as an antagonist of NMDA and kainate receptors. As FICZ can also be formed by photooxidation of Trp, it is most likely to act as antagonist of NMDA and possibly also kainate receptors.

Interpretation of our docking results of KA and QA to GABA receptors is however more complex. That KA docking to GABA_A may indicate an agonist behaviour is suggested by the ability of bicuculine (a GABA_A antagonist) to inhibit KA-enhanced swimming and climbing of mice in the forced-swim test [56]. That KA may also act as a GABA_B agonist is suggested by the similar behavior of KA produced by KMO inhibition [57] and the GABA_B agonist baclofen [58]: that of inhibition of drug-seeking. That KA does not act as a GABA antagonist is further suggested by a study in hippocampal slice preparations [59] and the slowing down of the decay kinetics of GABA_A-evoked currents in cultured neurons [60]. Depression of GABAergic neurotransmission can be mediated by activation of kainate receptors and, as a kainate receptor antagonist, KA has been reported to reduce the kainate-dependent disinhibition by synaptic glutamate release [61]. By contrast, activation of glutamate receptors by QA should exert the opposite effects on GABAergic neurotransmission, namely antagonism of both GABA_A and GABA_B receptors. That QA may act as a GABA_A and GABA_B antagonist is suggested by the increased density of GABA_A receptors following QA lesions in striatum [62] and the ability of baclofen to ameliorate the effects of QA on pyramidal neurons of the rat CA1 sector of dorsal hippocampus [63].

With lumirubin and FICZ, their potential effects on the above receptors have not been studied and should therefore be the subject of future investigations. FICZ is almost certain to be produced from Trp during NNH phototherapy. In adult plasma, Trp exists largely (90-95%) bound to albumin. The low albumin in the newborn infant results in increased free Trp availability. At birth, both free and total (free + albumin-bound) [Trp] in the newborn infant are double those in maternal circulation (see [64] and references therein). Light therapy is almost certain to convert some Trp into FICZ in infant tissues including the brain. FICZ possesses a range of properties [24]. The effects of FICZ on immune function are complex, multifactorial and dose-dependent. FICZ at low protective levels, as would be expected during phototherapy, can induce an antiinflammatory state [39] to counter bilirubin toxicity.

5. Conclusions and Comments

Exploring the formation of FICZ and lumirubin and their actions during phototherapy of NNH may represent a novel approach in understanding the pathophysiology and therapy of the newborn jaundice. The interactions between Trp and bilirubin metabolism and the potential mechanisms involved are illustrated in Figure 1. Whereas lumirubin may protect the newborn infant through its binding to the AhR, mainly by reducing inflammation [20], its potential effects on neuronal function remain to be explored. Studying the effects of lumirubin and FICZ on Trp metabolism along the KP in relation to immune and neuronal functions coupled with assessments of cytokine and chemokine levels and glutamatergic activity in experimental settings and where appropriate in the newborn infant should provide a platform for understanding the pathophysiology and neurological features of NNH and kernicterus and point towards pharmacological therapeutic options. Exploring IDO1 induction and increased KP metabolites in kernicterus may yield important information. The present molecular docking findings of direct interactions of KA and QA with GABA receptors are novel observations worthy of experimental scrutiny at the pharmacological and behavioral levels. The present study illustrates the value of molecular simulation in revealing potential biological and therapeutic mechanisms.

Conflict of interest

None declared

Funding

None

Supplementary File Link:

https://www.fortunejournals.com/supply/SupplyJPPCH-11353.pdf

References

1. Lathe GH, Walker M. The synthesis of bilirubin glucuronide in animal and human liver. Biochem J 70 (1958): 705-712.
2. Brites D. The evolving landscape of neurotoxicity by unconjugated bilirubin: role of glial cells and inflammation. Front Pharamcol 3 (2012): 88.
3. Badawy AA-B. Kynurenine pathway of tryptophan metabolism: regulatory and functional aspects. Int J Tryptophan Res 10 (2017): 1-20.
4. Fallarino F, Grohmann U, Vacca C, et al. T cell apoptosis by tryptophan catabolism. Cell Death Differ 9 (2002): 1069-1077.
5. Terness P, Bauer TM, Rose L, et al. Inhibition of allogeneic T cell proliferation by indoleamine 2,3-dioxygenase-expressing dendritic cells: mediation of suppression by tryptophan metabolites. J Exp Med 196 (2002): 447-457.
6. Wirthgen E, Hoeflich A, Rebl A, et al. Kynurenic acid: the Janus-faced role of an immunomodulatory tryptophan metabolite and its link to pathological conditions. Front Immunol 8 (2017): 1957.
7. Stone TW, Perkins MN. Quinolinic acid: a potent endogenous excitant at amino acid receptors in CNS. Eur J Pharmacol 72 (1981): 411-412.
8. Stone TW. The neuropharmacology of quinolinic and kynurenic acids. Pharmacol Rev 45 (1993): 309-385.
9. Hansen TWR, Wong RJ, Stevenson D. Molecular physiology and pathophysiology of bilirubin handling by the blood, liver, intestine, and brain in the newborn. Physiol Rev 100 (2020): 1291-1346.
10. Zunszain PA, Anacker C, Cattaneo A, et al. Interleukin-1β: A new regulator of the kynurenine pathway affecting human hippocampal neurogenesis. Neuropsychopharmacology 37 (2012): 939-949.
11. Usman F, Diala UM, Shapiro SM, et al. Acute bilirubin encephalopathy and its progression to kernicterus: current perspectives. Res. Reprod Neonatol 8 (2018): 33-44.
12. Lugo-Huitrón R, Muñiz PU, Pineda B, et al. Quinolinic acid: an endogenous neurotoxin with multiple targets, Oxid Med Cell Longevity 104024 (2013): 1-14.
13. Badawy AA-B. Multiple roles of haem in cystathionine β-synthase activity: implications for hemin and other therapies of acute hepatic porphyria, Biosci Rep 41 (2021): 1-24.
14. Abu-Bakar A, Arthur DM, Wikman ADS, et al. Metabolism of bilirubin by human cytochrome P450 2A6, Toxicol Appl Pharmacol 261 (2012): 50-58.
15. Jašprová J, Dal Ben M, Vianello E, et al. The biological effects of bilirubin photoisomers. PlosOne 11 (2016): e0148126.
16. Creeden JF, Gordon DM, Stec DE, Hinds TD. Bilirubin as a metabolic hormone: the physiological relevance of low levels, Am J Physiol: Endocrinol Metab 320 (2021): E191-E207.
17. Vreman HJ, Kourula S, Jašprová J, et al. The effect of light wavelength on in vitro bilirubin photodegradation and photoisomer production, Pediatric Res 85 (2019): 865-873.
18. Ebbesen F, Madsen PH, Vandborg PK, et al. Bilirubin isomer distribution in jaundiced neonates during phototherapy with LED light centered at 497 nm (turquoise) vs. 459 nm (blue). Pediatric Res 80 (2016): 511-515.
19. Knox I, Ennever JF, Speck WT. Urinary excretion of an isomer of bilirubin during phototherapy, Pediatric Res 19 (1985): 198-201.
20. Capková N, Pospíšilová V, Fedorová V, et al. The effects of bilirubin and lumirubin on the differentiation of human pluripotent cell-derived neural stem cells, Antioxidants 210 (2021): 1532.
21. Dvorák A, Pospíšilová K, Zízalová K, et al. The effects of bilirubin and lumirubin on metabolic and oxidative stress markers, Front Pharmacol 12 (2021): 567001.
22. Rannug A, Rannug U, Rosenkranz HS, et al. Certain photooxidized derivatives of tryptophan bind with very high affinity to the Ah receptor and are likely to be endogenous signal substances, J Biol Chem 262 (1987): 15422-15427.
23. Rannug A. 6-Formylindolo[3,2-b]carbazole, a potent ligand for the aryl hydrocarbon receptor produced both endogenously and by microorganisms can either promote or restrain inflammatory responses, Front Toxicol 4 (2022):
24. Dawood S, Badawy AA-B. Molecular docking of FICZ (6-formylindolo[3,2-b] carbazole) to kynurenine pathway enzymes: biological basis of a potential drug. J Pharmacol Pharmaceutics 1 (2024): 1-10. Art 1004.
25. Dawood S, Zarina S, Bano. Docking studies of antidepressants against single crystal structure of tryptophan 2, 3-dioxygenase using Molegro Virtual Docker software. Pakistan J Pharmaceut Sci 27 (2014): 1521-1539.
26. Dawood S, Bano S, Badawy AA-B. Inflammation and serotonin deficiency in major depressive disorder: molecular docking of antidepressant and antiinflammatory drugs to tryptophan and indoleamine 2,3 dioxygenases, Biosci Rep 42 (2022): 1-12.
27. Sakurai S, Shimizu T, Ohto U. The crystal structure of the AhRR-ARNT heterodimer reveals the structural basis of the repression of AhR-mediated transcription, J Biol Chem 292 (2017): 17609-17616.
28. Karakas E, Simorowski N, Furukara H. Structure of the zinc-bound amino terminal domain of the NMDA receptor NR2B subunit. Embo J 28 (2009): 3910-3920.
29. Miller PS, Aricescu AR. Crystal structure of a human gabaa receptor. Nature. 512 (2014): 270-275.
30. Geng Y, Bush M, Mosyak L, et al. Structural mechanism of ligand activation in human GABA(B) receptor, Nature 504 (2013): 254-259.
31. Unno M, Shinohara M, Takayama K, et al. Binding and selectivity of the marine toxin neodysiherbaine A and its synthetic analogues to GluK1 and GluK2 kainate receptors. J Molec Biol 413 (2011): 667-683.
32. Forouhar F, Anderson JL, Mowat CG, et al. Molecular insights into substrate recognition and catalysis by tryptophan 2,3-dioxygenase. Proc Natl Acad Sci USA 104 (2007): 473-478.
33. Crosignani S, Bingham P, Bottemanne P, et al. Discovery of a novel and selective indoleamine 2,3-dioxygenase (IDO-1) inhibitor 3-(5-Fluoro-1H- indol-3-yl)pyrrolidine-2,5-dione(EOS200271/PF06840003) and its characterization as a potential clinical candidate. J Medicinal Chem 60 (2017): 9617-9629.
34. Kim HT, Na BK, Chung J, et al. Structural basis for inhibitor- induced hydrogen peroxide production by kynurenine 3-monooxygenase. Cell Chem Biol 25 (2018): 426-438.
35. Nematollahi A, Sun G, Harrop SJ, et al. Structure of the PLP-form of the human kynurenine aminotransferase II in a novel spacegroup at 1.83 angstrom resolution. Int J Molec Sci 17 (2016): 446: 1-11.
36. Wong YK, Wood BS. Relative roles of phototherapy and phenobarbitone in treatment of nonhaemolytic neonatal jaundice. Arch Dis Childhood 48 (1973): 704-708.
37. Hubbard TD, Murray IA, Perdew GH. Indole and tryptophan metabolism: endogenous and dietary routes to Ah receptor activation. Drug Metab Dispos 43 (2015): 1522-1535.
38. Sinal CJ, Bend JR. Aryl hydrocarbon receptor-dependent induction of Cyp1a1 by bilirubin in mouse hepatoma Hepa 1c1c7 cells. Molec Pharmacol 5 (1997): 590-599.
39. Julliard W, Fechner JH, Mezrich JD. The aryl hydrocarbon receptor meets immunology: friend or foe? A little of both. Front Immunol 5 (2014): 458.
40. Rannug A, Ranng U. The tryptophan derivative 6-formylindolo[3,2-b)carbazole, FICZ, a dynamic mediator of endogenous aryl hydrocarbon receptor signaling, balances cell growth and differentiation. Crit Rev Toxicol 48 (2018): 555-574.
41. Ennever JF, Costarino AT, Polin RA, et al. Rapid clearance of a structural isomer of bilirubin during phototherapy. J Clin Invest 9 (1987): 1674-1678.
42. Jašprová J, Dal Ben M, Hurný D. Neuro-inflammatory effects of photodegradative products of bilirubin. Scientific Rep 8 (2018): 7444.
43. DiNatale BC, Murray IA, Schroeder JC, et al. Kynurenic acid is a potent endogenous aryl hydrocarbon receptor ligand that synergistically induces interleukin-6 in the presence of inflammatory signaling. Toxicol Sci 115 (2010): 89-97.
44. Banzola I, Mengus C, Wyler S, et al. Expression of indoleamine 2,3- dioxygenase induced by IFN-γ and TNF-α as potential biomarker of prostate cancer progression. Front Immunol 29 (2018): 1051.
45. Vogel CFAS, Goth SR, Dong B, et al. Aryl hydrocarbon receptor signaling mediates expression of indoleamine 2,3-dioxygenase. Biochem Biophys Res Commun 375 (2008): 331-335.
46. Litzenburger UM, Opitz CA, Sahm F, et al. Constitutive IDO expression in human cancer is sustained by an autocrine signaling loop involving IL-6, STAT3 and the AHR. Oncotarget 5 (2014): 1038-1051.
47. Dere E, Boverhof DR, Burgoon LD, et al. In vivo-in vitro toxicogenomic comparison of TCDD-elicited gene expression in Hepa1c1c7 mouse hepatoma cells and C57BL/6 hepatic tissue. BMC Genomic 7 (2006): 80.
48. Sadek A, Somarribas Patterson LF, Öztürk S, et al. IL4I1 is a metabolic immune checkpoint that activates the AhR and promotes tumor progression. Cell 182 (2020): 1252-`1270 e20-e34.
49. Ha HC, Snyder SH. Poly (ADP-ribose) polymerase is a mediator of necrotic cell death by ATP depletion. Proc Natl Acad Sci USA 96 (1999): 13978-13982.
50. Morales JC, Li L, Fattah FJ, et al. Review of poly (ADP-ribose) polymerase (PARP) mechanisms of action and rationale for targeting in cancer and other diseases. Crit Rev Eukaryotic Gene Expr 24 (2014): 15-28.
51. Tang K, Wu Y-H, Song Y, et al. Indoleamine 2,3-dioxygenase 1 (IDO1) inhibitors in clinical trials for cancer Immunotherapy. J Hematol Oncol 14 (2021): 68.
52. Badawy AA-B. Tryptophan metabolism and disposition in cancer biology and immunotherapy. Biosci Rep 42 (2022): BSR20221682.
53. Shin JH, Zhang L, Murillo-Sauca O, et al. Modulation of natural killer cell antitumor activity by the aryl hydrocarbon receptor. Proc Natl Acad Sci USA 110 (2013): 12391-12396.
54. Arabnezhad M-R, Montazeri-Najafabady N, Chatrabnous N, et al. Antiandrogenic effect of 6-formylindolo[3,2-b]carbazole (FICZ) in LNCaP cells is mediated by the aryl hydrocarbon-androgen receptors cross-talk. Steroids 153 (2020): 108508.
55. Fan C, Wang Q, van der Zon G, et al. OVOL1 inhibits breast cancer cell invasion by enhancing the degradation of TGF-β type I receptor. Signal Transd Targeted Ther 29 (2022): 126.
56. Tanaka M, Bohár Z, Martos D, et al. Antidepressant-like effects of kynurenic acid in a modified forced swim test. Pharmacol Rep 72 (2020): 449-55.
57. Vengeliene V, Cannella N, Takahashi T, et al. Metabolic shift of the kynurenine pathway impairs alcohol and cocaine seeking and relapse. Psychopharmacology 233 (2016): 3449-3459.
58. Cousins S, Roberts DCS, de Wit H. GABAB receptor agonists for the treatment of drug addiction: a review of recent findings. Drug Alcohol Depend 65 (2002): 209-220.
59. Stone TW. Kynurenines do not antagonise GABA in rat hippocampus. Eur J Pharmacol 128 (1986): 81-83.
60. Mok MHS, Fricker A-C, Weil A, et al. Electrophysiological characterisation of the actions of kynurenic acid at ligand-gated ion channels. Neuropharmacology 7 (2009): 242-249.
61. Min M-Y, Melyan Z, Kullmann DM. Synaptically released glutamate reduces g-aminobutyric acid (GABA) ergic inhibition in the hippocampus via kainate receptors. Proc Natl Acad Sci USA 96 (1999): 9932-9937.
62. Nicholson LF, Faull RL, Waldvogel HJ. GABA and GABAA receptor changes in the substantia nigra of the rat following quinolinic acid lesions in the striatum closely resemble Huntington's disease. Neuroscience 66 (1995): 507-521.
63. Beskid M, Rózycka Z, Taraszewska A. Quinolinic acid and GABA-B receptor ligand: effect on pyramidal neurons of the CA1 sector of rat's dorsal hippocampus following peripheral administration. Folia Neuropathologia 37 (1995): 99-106.
64. Badawy AA-B. The tryptophan utilization concept in pregnancy. Obstetric Gynecol Sci 57 (2014): 249-259.