In Silico Analysis of the Interaction of Avian Aryl Hydrocarbon

Feb 18, 2015 - In this study, the docking analysis of FICZ in avian AHRs revealed that there are a large number of amino acid residues responsible for...
1 downloads 11 Views 4MB Size
Download PDF
Article pubs.acs.org/est

In Silico Analysis of the Interaction of Avian Aryl Hydrocarbon Receptors and Dioxins to Decipher Isoform‑, Ligand‑, and SpeciesSpecific Activations Masashi Hirano,† Ji-Hee Hwang,‡ Hae-Jeong Park,‡ Su-Min Bak,‡ Hisato Iwata,† and Eun-Young Kim*,‡ †

Center for Marine Environmental Studies (CMES), Ehime University, Matsuyama 790-8577, Japan Department of Life and Nanopharmaceutical Science and Department of Biology, Kyung Hee University, Hoegi-Dong, Dongdaemun-Gu, Seoul 130-701, Korea

‡

S Supporting Information *

ABSTRACT: The aryl hydrocarbon receptor (AHR) mediates toxic responses to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and other dioxin-like compounds (DLCs). Avian species possess multiple AHR isoforms (AHR1, AHR1β, and AHR2) that exhibit species- and isoform-specific responses to ligands. To account for the ligand preference in terms of the structural features of avian AHRs, we generated in silico homology models of the ligand-binding domain of avian AHRs based on holo human HIF-2α (PDB entry 3H7W). Molecular docking simulations of TCDD and other DLCs with avian AHR1s and AHR2s using ASEDock indicated that the interaction energy increased with the number of substituted chlorine atoms in congeners, supporting AHR transactivation potencies and World Health Organization TCDD toxic equivalency factors of congeners. The potential interaction energies of an endogenous AHR ligand, 6-formylindolo [3,2-b] carbazole (FICZ) to avian AHRs were lower than those of TCDD, which was supported by a greater potency of FICZ for in vitro AHR-mediated transactivation than TCDD. The molecular dynamics simulation revealed that mean square displacements in Ile324 and Ser380 of TCDD-bound AHR1 of the chicken, the most sensitive species to TCDD, were smaller than those in other avian AHR1s, suggesting that the dynamic stability of these amino acid residues contribute to TCDD preference. For avian AHR2, the corresponding residues (Val/Ser or Val/Ala type) were not responsible for differential TCDD sensitivity. Application of the three-dimensional reference interaction site model showed that the stabilization of TCDD binding to avian AHRs may be due to the solvation effect depending on the characteristics of two amino acids corresponding to Ile324 and Ser380 in chicken AHR1. This study demonstrates that in silico simulations of AHRs and ligands could be used to predict isoform-, ligand-, and species-specific interactions.

■

INTRODUCTION Dioxin-like compounds (DLCs), including polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans (PCDD/Fs), and coplanar (dioxin-like) polychlorinated biphenyls (PCBs) are ubiquitously distributed in the ecosystem. Birds that belong to the upper trophic level accumulate high levels of DLCs through the food web.1,2 Chronic exposure to DLCs in birds elicits alteration of gene expression profiles, induction of xenobiotic metabolizing enzymes, developmental defects, and population decline.3−5 One of the molecular targets of DLCs is the aryl hydrocarbon receptor (AHR), which belongs to the basic helix−loop−helix/ Per−Arnt−Sim (bHLH/PAS) protein family.6 AHR, which is evolutionarily well conserved across invertebrate and vertebrate species,7 is a ligand-dependent transcription factor that mediates toxic effects induced by exposure to DLCs.8 The PAS domain of the AHR consists of two hydrophobic repeat regions, A and B, and PAS B functions as a ligand-binding domain (LBD).9,10 In the absence of ligands, AHR is present in the cytosol as a complex © 2015 American Chemical Society

with two chaperone HSP90 proteins, p23 and immunophilin-like protein XAP2. When ligands bind to the AHR LBD, the complex dissociates from AHR and the AHR subunit translocates into the nucleus. Following the dimerization with its heterodimerization partner, the AHR nuclear translocator (ARNT), AHR-ARNT binds to a specific DNA sequence, xenobiotic responsive element (XRE), in the promoter region of multiple genes, including cytochrome P450 1A (CYP1A), to regulate their expression.11 Although only a single AHR is present in mammals, at least two types of AHR isoforms (AHR1 and AHR2) have been found in fish and birds.12,13 Our recent study revealed that, in addition to AHR1 and AHR2, there is a third AHR isoform (AHR1β), which is phylogenetically close to AHR1, in the chicken.14 In vitro Received: Revised: Accepted: Published: 3795

July 31, 2014 February 15, 2015 February 18, 2015 February 18, 2015 DOI: 10.1021/es505733f Environ. Sci. Technol. 2015, 49, 3795−3804

Article

Environmental Science & Technology

Homology Modeling. Details are given in the Supporting Information. Molecular Docking of DLCs and FICZ to Avian AHR LBD Models. Molecular docking simulations of DLCs to avian AHR LBDs were carried out using the alpha sphere and excluded volume-based ligand−protein docking (ASEDock) program in the MOE.18,19 Details are given in Supporting Information. Reporter Gene Assay. The AHR-mediated transactivation potency of an endogenous AHR ligand, FICZ, was measured by the in vitro AHR reporter gene assay that we developed in our previous studies.4,5,12−14 Details of the assay are given in Supporting Information. MD Simulation. Initially, dummy atoms were removed from the final homology models of avian AHR LBDs prepared for ASEDock analyses. The most stable configuration of TCDD obtained from ASEDock was then manually docked into the active site of the AHR LBDs. The docked complex of AHR with TCDD was used as an initial structure for molecular dynamics (MD) simulation. The protonation state of the AHR LBD structure was adjusted to pH 7.0 in the Protonate 3D program. Periodic boundary conditions filled with water molecules were applied. A PFROSST force field was employed to optimize the structure. For the MD simulation, the Nose−Hoover−Andersen algorithm was used as a NVT ensemble. The simulation was subjected to 20 ps heating and equilibration from 100 to 300 K, followed by 1 ns simulation at 300 K at 1 fs time steps. Coordinate trajectories were recorded every 0.5 ps (total 2080 structures). The stability of the ligand−protein complex was evaluated by the root-mean-square deviation (RMSD). The resulting MD structures were energetically minimized, and the interaction energy was calculated. The cavity volumes for ligand binding pockets (LBPs) were calculated by the AtomRegion program based on alpha spheres obtained from the Site Finder module with a grid space of 0.1 Å. To monitor the behavior of amino acids, trajectories of the free MD simulations were analyzed using the Trajectory Tracer module. In addition, the mean square displacement (MSD) was calculated using the Trajectory Analysis module. To assess the solvation effect on TCDD binding in LBP, the 3D reference interaction site model (3D-RISM) theory was applied.20−23 The PFROSST force field was employed for the energy minimization and the solvation effect. The probability density of hydrophobic molecules was calculated for the analysis of solvation effect.

expressed proteins of all avian AHR isoforms showed TCDDbinding and XRE-driven transactivation potencies.12−14 Even though the AHR-mediated signaling pathway is highly conserved among vertebrates, there are large interspecies differences in sensitivity to TCDD and to other DLCs. In birds, positive correlations have been reported between in vitro relative transactivation potencies of DLCs mediated by AHR1s from certain species and embryonic mortalities of the same species in ovo administered with DLCs.15,16 This implies that AHR activation is a critical event that precedes the outcome of toxicity induced by DLCs in birds. These in ovo and in vitro experiments have shown that the chicken (Gallus gallus) is ranked as the most sensitive species; the black-footed albatross and ring-necked pheasant are moderately sensitive; and the common (great) cormorant and common tern are the least sensitive species. Nevertheless, the molecular basis for the differential sensitivity to DLCs in birds is not fully understood. The differential TCDD sensitivity in avian species has been explained by the differences in two amino acid residues in the LBD of avian AHR1. AHR1 of the chicken (ckAHR1) has Ile324 and Ser380 in the LBD, AHR1 of the black-footed albatross (bfaAHR1) has Ile325 and Ala381, and AHR1 of the common cormorant (ccAHR1) has Val325 and Ala381 in the corresponding sites, showing TCDD-EC50 values for AHR1-mediated transactivation in the order of ckAHR1 (0.030 nM) < bfaAHR1 (0.077 nM) < ccAHR1 (0.36 nM). Conversely, the “two amino acids” rule was less valid for avian AHR2. Although the former amino acid residue in ckAHR1 is substituted for Val302 in the corresponding site of ckAHR2, and both residues are substituted for Val318−Ala374 in bfaAHR2 and ccAHR2, ckAHR2 is nonresponsive to TCDD13,17 and ccAHR2 is more sensitive to TCDD than bfaAHR2.4,5,12 Hence, we hypothesized that other sites may be critical for the AHR2mediated transactivation potency by DLCs; however, some things remain unknown. First, it is not clear why the two amino acid residues in the LBD are critical for the interaction of TCDD−AHR1 and not for TCDD−AHR2. Second, it is unknown why the Ile−Ser type of AHR1 is more sensitive to TCDD than other types of AHR1s, i.e., Ile−Ala and Val−Ala. To understand the molecular basis of the species-, ligand-, and AHR isoform-specific responses, the present study investigated the structural characteristics of the interaction of multiple avian AHR isoforms with TCDD and other DLCs by in silico docking analyses. For comparison, the interaction of an endogenous AHR ligand, 6-formylindolo [3,2-b] carbazole (FICZ) with avian AHRs was also analyzed. We initially tested the availability and limitations of in silico docking models as a tool to screen potential ligands of avian AHRs by comparing the ligand−AHR interaction energy with the transactivation potency of each ligand in in vitro AHR assays. We then sought to determine the role of the two amino acid residues in ligand interaction using a trajectory analysis module and three-dimensional (3D) reference interaction site model in the molecular dynamics simulation.

■

RESULTS AND DISCUSSION Homology Models of Avian AHR LBDs. In this study, we applied the crystal structure of the PAS-B domain of HIF-2α as the template for constructing avian AHR homology models. Because no crystal structure of AHR PAS-B domain has been experimentally determined, most AHR models have been built based on the crystal structure of the PAS-B domain of HIF2α,24−31 of which the sequence among proteins with structural data showed the highest identity to the sequence of AHR PAS-B domain. These previous studies have indicated that the homology model of AHR PAS-B domain is useful to account for the species differences in TCDD binding and to predict the potential binding ability of AHR ligands. Moreover, the function of AHR PAS-B domain estimated from in silico analysis has been proven to be supportive by several in vitro assays.25,26,28 Because the template itself may affect the accuracy of homology models, we initially constructed models of the ckAHR1 LBD derived from two different templates; one

■

METHODS Data Set and Template Preparation. The construction of homology models of avian AHR LBDs and the in silico docking analysis of the interaction potential with ligands were performed using Molecular Operating Environment (ver. MOE 2012.10; Chemical Computing Group Inc., Montreal, QB, Canada) software package. Details of the in silico docking analysis are given in Supporting Information. 3796

DOI: 10.1021/es505733f Environ. Sci. Technol. 2015, 49, 3795−3804

Article

Environmental Science & Technology reference model was based on the crystal structure of apo huHIF2α, and the other was on that of holo huHIF-2α. The properties of the two models are summarized in Table S1 in Supporting Information. In the Ramachandran plots, φ/ψ torsion angles were within the favorable region for each ckAHR1 LBD model (Figure S1A,B in Supporting Information). The aberrational residues in the stereochemistry and the clash of atoms were not observed. RMSD values of the Cα atoms for the superimposition of ckAHR1 homology models with the corresponding apo and holo huHIF-2α were 0.87 and 1.09 Å, respectively (Table S1 in Supporting Information), indicating high structural similarities of both AHR models with the respective huHIF-2α templates. Meanwhile, when compared between the two homology models, the RMSD value was 2.47 Å, showing a structural difference between the models. The volumes in the main cavity of ckAHR1 LBDs were then calculated for the two structural models. The cavity volumes of the representative apo and holo ckAHR1 homology models were 99.6 and 308 Å3, respectively. Again, these results indicate significant conformational differences between the homology models constructed based on the apo and holo huHIF-2α template structures. In addition, we measured the interaction energies for TCDD binding using ASEDock program. The U_dock values of TCDD binding were 273 and −8.75 kcal/mol for apo and holo ckAHR1 homology models, respectively (Table S1 in Supporting Information). The higher value in the apo huHIF-2α-based AHR form is probably due to a smaller LBP in the LBD. Thus, the holo template may be more suitable for constructing a homology model of avian AHR LBDs than the apo template. There is no convincing findings that apo huHIF-2α-based AHR homology model is more useful than holo huHIF-2α-based model. Many papers have shown great achievements using AHR homology models constructed based on the holo huHIF-2α structure as a template rather than the apo structure. Motto et al.26 clearly stated the use of the holo huHIF-2α structure as the template to enhance the accuracy of the binding site for molecular docking applications. In addition, holo huHIF-2αbased human AHR model constructed by Xing et al.27 succeeded in predicting the rank of binding affinity of TCDD and other DLCs and the rank of their toxicity. In more recent papers,30,31 their AHR models were constructed by using the holo HIF-2α structure as the template to predict the binding modes. Based on these and our data, we chose to apply the holo huHIF-2α structure (PDB entry 3H7W) as a template to construct the avian AHR LBD models for further docking analyses, if not otherwise specified. Amino acid sequences of the regions modeled using the template 3H7W were aligned (Figure S2 in Supporting Information). The avian AHR PAS-B domains shared 22−32% identity with huHIF-2α (Figure S3 in Supporting Information). For each avian AHR LBD, 125 candidates of 3D homology model were generated from the 3H7W template. Among the candidates, the top-ranked structure was sorted by total potential energy based on GB/VI scoring. To assess the validity of the topranked model, docking of TCDD with the top three ranked homology models in ckAHR1, -1β, and -2 was simulated. Results indicated that there were only small differences in the interaction energies of TCDD among the top three ranked homology models on any of AHR isoforms (Table S2 in Supporting Information). These results clearly demonstrated the utility of the top-scored model we used in this study. Thus, the top ranked model was used as the final model structure of each AHR LBD (Figure S4 in Supporting Information). Although the protein

structure for each model varied for a few amino acid residues in the loop region, most were positioned in the sterically allowed region of the Ramachandran plot (see the representative plots in Figure S1B−H in Supporting Information). Furthermore, this high degree of structural conservation was confirmed by no atom clashes and low RMSD values between each model and template structure (0.73−0.91 Å on the Cα atoms). These results demonstrate the high accuracy of these homology models of avian AHRs. Previous studies reported that the structures of avian AHR1-, 1β-, and 2-LBDs are composed of five or six β-sheets and a central helix.28,29 High stereochemical similarities were observed between chicken, albatross, and cormorant AHR structures, showing 0.57−1.18 Å (avg 0.89 Å) of RMSD values (Figure S5 in Supporting Information). These findings suggest that avian AHR LBDs have highly homologous structures. On the other hand, RMSD values in comparison with ckAHR1β were higher (0.95− 1.18 Å), implying that ckAHR1β has a slightly different structure compared with other AHR LBDs. Potential LBPs in AHR LBDs were identified based on alpha spheres gained using Site Finder,18 and the volume of LBPs was measured by AtomRegion. The estimated cavity volumes of LBPs were 308, 420, 309, and 376 Å3 for ckAHR1, ckAHR1β, bfaAHR1, and ccAHR1, respectively (Table S3 in Supporting Information). For ck-, bfa-, and ccAHR2 LBD models, the LBP volumes were 398, 414, and 340 Å3, respectively (Table S3). It has been shown that two amino acids located in the avian AHR1 LBD, corresponding to residues 324 and 380 within the ckAHR1 LBD, contribute to the internal cavity size and consequently affect TCDD binding affinity in birds.28,29,32 In fact, in silico mutagenesis analysis revealed that each ckAHR1 LBD mutant model (Ile324Val and Ser380Ala) artificially enlarged the cavity volume. Likewise, a mutant of both amino acids (I324 V_S380A) further increased the volume.29 Our results indicated that the volume of ckAHR1 (Ile/Ser type) was comparable to that of bfaAHR1 (Ile/Ala type), whereas these volumes were both smaller than that in ccAHR1 (Val/Ala). Additionally, the cavity volumes of ck- and bfaAHR2 (Val/Ser or Val/Ala type) were 30% larger than those of ck- and bfaAHR1 (Table S3), implying that the amino acid at residue 324 significantly contributes to the cavity volume in avian AHR isoforms. This is also supported by the similar cavity volumes of both ccAHR1 and 2, which are of the Val/Ala type. Among the tested avian AHR LBDs, ckAHR1β showed the greatest cavity volume. Because ckAHR1β has Ile315 and Val371 (Ile/Val type), this isoform is predicted to have a smaller volume compared with the Ile/Ala type, such as bfaAHR1.24,25,29 However, the volume of ckAHR1β was larger than that of bfaAHR1. The amino acid sequences of the PAS-B domain were highly conserved in ck-, bfa-, and ccAHR1, sharing more than 98% identity, whereas ckAHR1β had 76−77% amino acid identities with 16 unmatched amino acids compared with other avian AHR1s (Figure S3 in Supporting Information). The difference in these unmatched amino acid sequences of the PASB domain may expand the volume of LBP. Molecular Docking Simulation of DLCs and FICZ. To evaluate the potential risk of a mixture of DLCs, a concept that incorporates toxic equivalency factor (TEF) of each DLC congener has been established. The TEFs are derived from the relative potency (REP) of each congener to a reference congener, TCDD, for AHR-mediated toxic effects. The TEFs are provided for mammals, birds, and fish by the World Health Organization (WHO) working group. The WHO avian-specific TEFs were 3797

DOI: 10.1021/es505733f Environ. Sci. Technol. 2015, 49, 3795−3804

Article

Environmental Science & Technology

Figure 1. Relationships between potential DLC-AHR interaction energies and WHO-TEFs: (A) ckAHR1, (B) ckAHR1β, (C) bfaAHR1, (D) ccAHR1, (E) ckAHR2, (F) bfaAHR2, and (G) ccAHR2. U_dock value (kilocalories per mole) as a potential DLC-AHR interaction energy was obtained from ASEDock analysis. TEFs for avian species were cited from Van den Berg et al.15 r- and p-values represent the Spearman’s correlation coefficient and the level of significance, respectively.

Table 1. Parameters of AHR Activation by TCDD Exposure and the Volume of AHR Ligand Binding Pockets ckAHR1 ckAHR1β bfaAHR1 ccAHR1 ckAHR2 bfaAHR2 ccAHR2

Bmaxa (fmol)

Kd (nM)a

EC50 (nM)b

U_dock (kcal/mol)

Eprot-lig (kcal/mol)c

vol (Å3)d

120 − 84.5 121 − 19.4 −

0.95 − 2.8 19.4 − 2.5 −

0.0047 0.041 0.077 0.36 − − 0.23

−8.8 −15.5 −9.9 −4.9 −12.5 −12.8 −3.6

−234 −93.8 −132 −88.8 −104 −97.3 −88.4

353 524 528 394 615 528 486

a

Data from [3H]TCDD binding to each AHR by sucrose velocity sedimentation assays.12,41 bData from in vitro AHR-mediated reporter gene assays.4,5,13,14 cInteraction energy of TCDD and AHR obtained by the MD simulation in this study. dVolume of the ligand-binding pocket of each AHR estimated by the MD simulation in this study. Grad space: 0.1 Å.

may be associated with the earlier findings that this AHR has no transactivation potency by TCDD treatment and is rather constitutively active.13 A lower U_dock value was observed from the docking simulation of ckAHR1β-TCDD when compared to that of ckAHR1-TCDD. Our previous study demonstrated that although in vitro expressed ckAHR1β has a binding potential to [3H]-labeled TCDD, the EC50 value for the ckAHR1β-mediated transactivation potency of TCDD was 13-fold higher than that for ckAHR1.14 This suggests that the AHR docking simulation can estimate the REPs of DLC congeners to a certain AHR but cannot account for the interisoform difference (Table S4 in Supporting Information). In terms of interspecies, U_dock values of bfaAHR1- and ccAHR1-TCDD were lower than that of ckAHR1-TCDD; this is not consistent with relative binding affinities observed in vitro (Table 1). Therefore, it is difficult to completely estimate interspecies differences from U_dock values. On the other hand, the simulations using apo-based avian AHR models showed that some DLCs failed to fit into the active site of the apo avian AHR models because the energyminimization calculation values exceeded the energy threshold. Moreover, all interaction energies of the 17 DLCs with the apo avian AHR models showed positive values, although the interaction energies of DLCs with these AHR models were significantly correlated with TEFs in most cases (Figure S7 in Supporting Information). These results suggest that the

assigned based on evidence from in vivo and in vitro studies, mostly on AHR-mediated toxicity in the chicken.15,33 To elucidate the AHR binding potency of each DLC congener, and to determine whether the REPs/TEFs can be predicted from in silico AHR binding potency, the docking simulation of each DLC to avian AHRs was performed using ASEDock. In this study, U_dock values that indicate the interaction energies of 17 DLC congeners with each avian AHR isoform were estimated (Table S4 in Supporting Information). In addition, variations of the top four scored conformations of TCDD in avian AHR models were assessed by evaluating RMSD values of TCDD. The results showed small RMSD values for all avian AHR models, suggesting the favorable accuracy of docking simulation and a stable conformation of TCDD in avian AHRs (Figure S6 in Supporting Information). For all AHR models constructed based on the holo huHIF-2α template, the interaction energies of 2,3,7,8-TCDD and 2,3,7,8-TCDF showed minus values lower than those of other congeners, while OctaCDD exhibited positive values. The ranking of binding affinities for each DLC congener estimated from the interaction energy was TCDD/F > PeCDD/F > HxCDD/F > HpCDD/F > OctaCDD/F. This indicates that U_dock values have a tendency to increase with the number of chlorine atoms in the congeners, except in the case of ckAHR2. This tendency is consistent with that of WHO avian TEFs (Figure 1). As for the case of ckAHR2, the poor tendency 3798

DOI: 10.1021/es505733f Environ. Sci. Technol. 2015, 49, 3795−3804

Article

Environmental Science & Technology

tally confirm AHR-mediated transactivation potency of FICZ, EC50 values were determined from the concentration−response curves from the in vitro reporter gene assay using ckAHR1 (Ile/ Ser type)12 and ccAHR1 (Val/Ala type)5 expression constructs. In ckAHR1 assay, EC50 values of FICZ and TCDD were found to be 0.008 and 0.07 nM, respectively. The EC50 value of FICZ for ccAHR1 was 0.018 nM, and the value of TCDD was 1.36 nM. Thus, avian AHR1 was more sensitive to FICZ than TCDD in our in vitro assay system, irrespective of the AHR genotype (Figure S10 in Supporting Information). This in vitro result is consistent with the U_dock value obtained in this study and supports the findings by Farmahin et al.;38 FICZ is a stronger inducer of AHR1-mediated luciferase activity than TCDD, and the two amino acid residues that affect sensitivity to DLCs (e.g., Ile/Ser in ckAHR1) in birds have no influence on the sensitivity to FICZ. In this study, the docking analysis of FICZ in avian AHRs revealed that there are a large number of amino acid residues responsible for the interaction with FICZ, and the amino acid residues located within 4.5 Å of FICZ in the best docking position are mostly similar to those of TCDD (Tables S5 and S6 in Supporting Information). Several AHR homology models have proposed that His291, Ser365, and Gln383 in human AHR;24,39 His285, Gly315, Ser359, and Gln377 in mouse AHR;24,37 and Ala354 and Ala370 in African clawed frog AHR1β40 (corresponding to Ser359 and Ala375 in mouse AHR, respectively) may be involved in FICZ binding. Our mouse AHR model showed that His285 interacted with the indole NH of FICZ through hydrogen bonding and that Ser359 and Gln377 participated in the reduction of solvent-exposed surface area (Figure S9 in Supporting Information), sharing critical amino acids with human, mouse, and frog AHRs (Table S6 in Supporting Information). For all avian AHR models, amino acid residues corresponding to His290, Ser380, and Gln382 in ckAHR1 were positioned within the 4.5 Å around the docked FICZ, and two residues corresponding to Ser380 and Gln382 were involved in FICZ binding by forming the indole NH hydrogen bonding or CH/π hydrogen bonding in ckAHR1, -1β, -2, and ccAHR1 (Table S6 in Supporting Information). As observed in the interaction with TCDD, Phe284 in ckAHR1 formed CH/π interaction with FICZ. Moreover, the amino acid residues corresponding to His290, Phe294, and Ser380 in ckAHR1 participated in the reduction of solvent-exposed surface areas in avian AHR1s (Figure S9 in Supporting Information), suggesting the contribution of these residues to FICZ binding. In AHR2, the CH/π interaction, which was not observed for the interaction with TCDD, was observed for the interaction with FICZ (Figure S9 in Supporting Information), implying that this endogenous ligand has a potential to stably bind to AHR2 as well as AHR1. MD Simulation for Avian AHR LBDs and TCDD. It was recently suggested that two amino acids corresponding to residues 324 and 380 in ckAHR1 are useful for predicting the dioxin sensitivity for in vitro AHR-mediated transactivation potency, as well as in ovo toxicity in avian species.29 The authors suggested that avian AHRs are classified into three types, type I (ckAHR1 type with Ile/Ser), type II (ring-necked pheasant AHR1 type with Ile/Ala), and type III (Japanese quail AHR1 type with Val/Ala). On the basis of the in vitro reporter gene assay using mutated AHR1 constructs, different avian species with the same type of AHR1 showed a similar sensitivity, and type I AHR1 was more sensitive than type II and type III. Our recent study provided evidence that ckAHR1β (a novel Ile/Val type) is less sensitive than type I.14 To verify whether the

interaction energies may predict the relative potency of individual DLCs but did not reflect the real stable interaction with low chlorinated DLCs including TCDD and TCDF because of steric hindrance in our apo AHR models. The results of these docking simulations indicate again that holo-based AHR models are superior to apo-based AHR models. Several comparative experimental and simulation studies have reported key amino acid residues involved in TCDD binding in mammals.24,25,34−36 A mouse AHR model showed that 26 amino acids are located within the 5 Å shell around the docked TCDD.26 On the basis of experimental mutagenesis and functional analysis, seven of these amino acids (Thr283, His285, Phe289, Tyr316, Ile319, Phe345, and Ala375) in the AHR ligand binding cavity were identified as a “TCDD bindingfingerprint” critical for optimal high TCDD binding affinity.25 Other studies revealed the importance of Phe318, Cys327, and Gln377 in the mouse AHR.35−37 In human AHR, the polar residues His291, Ser365, and Gln383 (His285, Ser359, and Gln377 in mAHR, respectively) contribute to TCDD binding by forming a hydrogen bond.24,31,38 The binding modes predicted for mouse and avian AHRs in this study are shown in Figure S8 in Supporting Information. Our docking models showed that 18− 22 amino acid residues in avian AHR1 and 2 were located within 4.5 Å of the ligand, and the amino acids surrounding TCDD were similar between the mammalian and avian AHRs (Table S5 in Supporting Information). In our mouse AHR model, TCDD formed two hydrogen bonds with Cys327 and His331 and CH/π interaction with Gln377, sharing critical amino acid residues with mouse AHR models in earlier studies. As for the ckAHR1 model, TCDD formed two hydrogen bonds with Cys299 and Cys332 and CH/π interaction with Phe284, as predicted in our mouse AHR model. Regarding ckAHR1β, Cys323 corresponding to Cys332 in ckAHR1 was involved in a hydrogen bond with TCDD, and CH/π interaction with Gln373 corresponding to mouse AHR Gln377 was observed. In bfaAHR1 and ccAHR1, a hydrogen bond was formed with Cys300 (corresponding to Cys299 in ckAHR1) and Met348, respectively. In the close vicinity of TCDD in both AHR1 models, three polar residues corresponding to His291, Ser365, and Gln383 in human AHR and two amino acid residues corresponding to Ile324 and Ser380 in ckAHR1 were allocated but did not participate in hydrogen bond interaction with TCDD in avian AHR1 models. However, these residues were involved in reducing solvent-exposed surface area. For avian AHR2 models, no noncovalent interactions with TCDD were observed, suggesting that this different binding mode may be responsible for the low transactivation potency by TCDD compared with avian AHR1.12,13 Taken together, these results indicate that the TCDD preference is mostly conserved in AHR orthologs but not in paralogues in avian species. For comparison with the result of TCDD docking analyses, we performed a docking simulation of FICZ, a tryptophan photoproduct known to be an endogenous AHR ligand. Similar to the simulation of TCDD, we initially simulated docking of FICZ with the top three ranked homology models in ckAHR1, -1β, and -2 to assess the validity of the top ranked model. Results showed only small differences in the interaction energies of TCDD among the models on any of AHR isoforms (Table S2 in Supporting Information). Thus, we used the top ranked model as the final model structure of each AHR LBD. The docking pose and the list of residues are shown in Figure S9 and Table S6 in Supporting Information, respectively. Interestingly, U_dock values of FICZ to avian AHRs were much lower than those of TCDD (Table S4 in Supporting Information). To experimen3799

DOI: 10.1021/es505733f Environ. Sci. Technol. 2015, 49, 3795−3804

Article

Environmental Science & Technology

Figure 2. MSD values of two amino acid residues in avian AHRs during 1 ns of the MD simulation: (A) ckAHR1, (B) ckAHR1β, (C) bfaAHR1, (D) ccAHR1, (E) ckAHR2, (F) bfaAHR2, (G) ccAHR2, and (H) mouse AHR. Isoleucine, green; serine, yellow; valine, cyan; alanine, magenta. Values in each figure are given as mean ± SD of MSD for each amino acid residue during 1 ns.

two variables. Pearson’s correlation analysis revealed a positive relationship between the interaction energy and the TCDD-EC50 in our in vitro AHR reporter gene assay (r = 0.83, p = 0.041) (Figure S11 in Supporting Information). Therefore, the potential interaction energy may be used as an index to predict AHRmediated responses of TCDD in avian species. Furthermore, it would also be able to evaluate the interspecies differences in sensitivity to TCDD. The cavity volume of ckAHR1 LBP estimated by the MD simulation (353 Å3, Table 1) was larger than the estimate for the LBP of ckAHR1 homology model constructed using apo huHIF-2α as a template (99.6 Å3, Table S1 in Supporting Information). This result indicates a clear ligand induced-fit effect on the cavity volume of LBP by the MD simulation. The characterization of atomic movement in proteins is essential for understanding their biological relevance. In this study, the MSD of atomic positions in amino acids residues corresponding to Ile324 and Ser380 in ckAHR1 were measured over 1 ns by the MD simulation. MSD is an indicator of atomic behavior, and the low value indicates a stable interaction with ligands. In this simulation, we measured the MSD, focusing on the side-chain atoms closest to TCDD of the two amino acid

different responses to TCDD depending on AHR genotypes are accounted for by tracing the trajectory of the respective two amino acids in each AHR, we conducted the MD simulation of the avian AHR−TCDD complex, and the kinetic stability was investigated. The highest ranked configuration for TCDD obtained by ASEDock analysis was further investigated up to 1 ns by MD simulation. During the MD simulation, the interaction energy reached an equilibrium state after 300 ps in each AHR (data not shown). The equilibrated potential interaction energies were estimated to be −234, −132, −93.8, and −88.8 kcal/mol for ckAHR1, bfaAHR1, ckAHR1β, and ccAHR1, respectively (Table 1). This trend is consistent with the trend of sensitivity to TCDD in each AHR1 as a consequence of variation at amino acid residues corresponding to 324 and 380 in ckAHR1,29 and with that of transactivation potency (TCDD-EC50) mediated by in vitro expressed AHR1 of TCDD (Table 1).4,5,12,13,41 As for AHR2, the potential energies were estimated to be −104, −97.3, and −88.4 kcal/mol for ckAHR2, bfaAHR2, and ccAHR2, respectively. To assess whether the TCDD−AHR interaction energy could be used as a tool to predict the AHR-mediated response to TCDD, we examined the relationship between these 3800

DOI: 10.1021/es505733f Environ. Sci. Technol. 2015, 49, 3795−3804

Article

Environmental Science & Technology

Figure 3. Analysis of solvation effects on TCDD binding in avian AHR LBDs. Desolvation penalty of (A) ckAHR1, (B) ckAHR1β, (C) bfaAHR1, (D) ccAHR1, (E) ckAHR2, (F) bfaAHR2, (G) ccAHR2, and (H) mouse AHR were analyzed by 3D-RISM. TCDD is colored in pink, and amino acid residues in each AHR are in gold. Ligands and water molecules are shown as a ball and stick model. ΔGbinding,solvation, the solvation free energy, is mapped for each AHR. Green surface represents the area where ΔGbinding,solvation ≤ −0.2 kcal/mol/ Å3, and red surface represents the area where ΔGbinding,solvation ≥ 0.2 kcal/mol/ Å3.

bfaAHR1 (Ile), and ccAHR1 (Val) were 0.15, 0.15, 0.17, and 0.64 Å2, respectively (Figure 2A−D). For the second amino acid residue in AHR1, the average MSD values were 0.11, 0.37, 0.12, and 0.17 Å2 for ckAHR1 (Ser), ckAHR1β (Val), bfaAHR1 (Ala),

residues corresponding to residues 324th and 380th in ckAHR1. In addition, we measured the MSD of the respective two amino acid residues in mouse AHR. For the first amino acid residue in AHR1, the average MSD values in ckAHR1 (Ile), ckAHR1β (Ile), 3801

DOI: 10.1021/es505733f Environ. Sci. Technol. 2015, 49, 3795−3804

Article

Environmental Science & Technology and ccAHR1 (Ala), respectively (Figure 2A−D). As for avian AHR2, at the first amino acid residue (Val in all), the average MSD value in ckAHR2, bfaAHR2, and ccAHR2 were 0.63, 0.33, and 0.26 Å2, respectively (Figure 2E−G). At the second position, the average values were 0.03, 0.07, and 0.08 Å2 for ckAHR2 (Ser), bfaAHR2 (Ala), and ccAHR2 (Ala), respectively (Figure 2E−G). In mouse AHR, the average MSD values of Ile319 and Ala375 were 0.15 and 0.10 Å2, respectively (Figure 2H). The motions of Ile319 and Ala375 in mouse AHR were similar to that of bfaAHR1 with the same Ile/Ala type, giving a moderate stability for the interaction with TCDD. With regard to the first amino acid residue in AHR1 and AHR2, the movement of Ile was lower than that of Val, suggesting that Ile provided greater stability for TCDD binding. It has been implied that mouse Ile319 and chicken Ile324 may stabilize the interaction with TCDD by providing a larger hydrophobic space because Ile has a longer side chain than Val.24,28 At the second position, Ser and Ala showed lower MSD values, whereas the motion of Val in ckAHR1β was greater (average MSD: 0.37 ± 0.16 Å2). It is known that the Val in DBA mouse and human AHRs leads to a low affinity for TCDD because it introduces a steric hindrance.28,35 Moreover, the MSD value of Ser359 in mouse AHR, which have been proven to be critical for TCDD binding in experimental mutagenesis analysis,26 showed the lowest value (Figure 2H), supporting that this amino acid had an important influence on TCDD binding. Overall, total MSDs from Ile324 and Ser380 in TCDDbound ckAHR1 were lower than MSDs from the different combinations in other avian AHRs. Thus, this trajectory analysis accounted for the findings in earlier in vitro studies. To examine the role of Ser380 in ckAHR1 and the respective amino acids in other AHRs, the solvation effects were examined using 3D-RISM, which has been applied to assess the effect of solvents on hydrogen bonding and the augmentation of ligand binding affinity by the “solvent-displacement effect”.42 Solvation geometry for TCDD binding was mapped in each avian AHR1 and 2 LBD (Figure 3). We initially predicted the location of water molecules based on the probability density distribution. Results showed that the isosurface of the probability density of water was consistent with the water molecule sterically allocated by MD simulation (Figure S12 in Supporting Information). We then evaluated the desolvation energy (ΔGbinding,solvation). The binding desolvation penalty maps showed that Val371 in ckAHR1β restricts the molecular interaction with TCDD, supporting the greater MSD value (Figure 3B). Furthermore, Ala381 in ccAHR1 contributed unfavorably to ΔGbinding,solvation, as indicated by the red surface, ΔGbinding,solvation ≥ 0.2 kcal/mol/Å3 (Figure 3D). On the other hand, Ser380 in ckAHR1 and Ala381 in bfaAHR1, as with Ser359 in mouse AHR, gave favorable solvation effects (negative desolvation energy) as represented by the green surface, ΔGbinding,solvation ≤ −0.2 kcal/mol/Å3 (Figure 3A,C,H). Moreover, for Ser380 of ckAHR1, a bridge effect via water molecules was observed (Figure 3A). However, this was not observed for Ala381 of bfaAHR1 (Figure 3C) because Ser is a polar residue and Ala is not. Thus, Ser380 may enhance the binding affinity to TCDD. This is consistent with a previous report indicating that Ser380 mutation led to a lower sensitivity to TCDD in an in vitro AHR-mediated reporter assay.32 As for AHR2, Ser358 in ckAHR2 and Ala374 in bfa- and ccAHR2 contributed favorably to the desolvation energy as indicated by the green surface (Figure 3E−G). In contrast, Val302, Val317, and Val317 in ck-, bfa-, and ccAHR2, respectively, weakened the stability to TCDD. In addition, the

region around the Val residue showed a low probability density of waters (Figure S12E−G in Supporting Information), and this observation was associated with a low solvation effect as represented by the red surface (Figure 3E−G). This supports the larger MSD value obtained in the MD simulation (Figure 2). Our previous studies indicated that ckAHR2 had no TCDD dose-dependent transactivation potency, and bfaAHR2 and ccAHR2 showed TCDD dose-dependency but appeared to have reduced transcriptional efficacies compared to the paralogous AHR1s.4,5,12,13 Therefore, AHR2-mediated transcriptional activities may be independent of the two variable amino acid residues (Val/Ser or Ala). Overall, these results suggest that the stabilization of TCDD binding to avian AHRs is due to the solvation effect depending on the characteristics of amino acids. In summary, to investigate the structural feature of avian AHRs, we constructed homology models of chicken, albatross, and cormorant AHR LBDs using the crystal structure of holo HIF-2α PAS B domain as a template. Molecular docking simulation of TCDD with avian AHR LBDs supported our previous findings that all tested avian AHR isoforms are capable of binding to TCDD, with negative interaction energy (U_dock) values. In addition, in silico binding affinity of DLCs based on the interaction energies was able to predict the rank of WHOassigned avian TEFs of DLCs. MD simulation roughly accounted for the interspecies differences in the in vitro AHR-mediated responses to TCDD exposure. MSDs in Ile324 and Ser380 of TCDD-bound ckAHR1 were lower than those in the corresponding sites (Val and Ala/Val) of other AHRs. This suggests that these two amino acids (Ile and Ser) contribute to TCDD preference, and the solvation effects of them relate to TCDD binding. Taken together, this study demonstrated that the structural difference of two amino acid residues in avian AHR homology models can be translated into different geometries of binding sites and binding modes.

■

ASSOCIATED CONTENT

S Supporting Information *

Data set and template preparation (Method S1); homology modeling (Method S2); molecular docking of DLCs to avian AHR LBD models (Method S3); reporter gene assay (Method S4); structural properties of homology modeling of ckAHR1 constructed using apo and holo HIF-2α templates (Table S1); validation of the top three scored homology models in ckAHR1, −1β, and −2 (Table S2); two amino acid residues of avian AHRs critical for TCDD binding and the estimated volume of the ligand-binding pocket of AHR homology models (Table S3); WHO-TEFs designated for avian species and U_dock values (kilocalories per mole) obtained from molecular docking simulation of DLCs to each avian AHR model using ASEDock program (Table S4); amino acid residues predicted to participate in TCDD binding in mouse and avian AHRs (Table S5); amino acid residues predicted to participate in FICZ binding in mouse and avian AHRs (Table S6); Ramachandran plots of avian AHR homology models (Figure S1); alignment of amino acid sequences of avian AHR LBDs and human HIF2α (Figure S2); amino acid sequence identities of avian AHR1s (Figure S3); in silico top ranked homology models of AHR1 and AHR2 LBDs from the chicken, black-footed albatross, and common (great) cormorant (Figure S4); RMSD values of avian AHR homology models (Figure S5); superposition of top four scored TCDD conformations obtained from ASEDock (Figure S6); relationships between potential interaction energies of DLCs with apobased AHR models and WHO-TEFs (Figure S7); binding modes 3802

DOI: 10.1021/es505733f Environ. Sci. Technol. 2015, 49, 3795−3804

Article

Environmental Science & Technology

(10) Denison, M. S.; Nagy, S. R. Activation of the aryl hydrocarbon receptor by structurally diverse exogenous and endogenous chemicals. Annu. Rev. Pharmacol. Toxicol. 2003, 43, 309−334. (11) Mandal, P. Dioxin: A review of its environmental effects and its aryl hydrocarbon receptor biology. J. Comp. Physiol., B 2005, 175 (4), 221−230. (12) Yasui, T.; Kim, E. Y.; Iwata, H.; Franks, D. G.; Karchner, S. I.; Hahn, M. E.; Tanabe, S. Functional characterization and evolutionary history of two aryl hydrocarbon receptor isoforms (AhR1 and AhR2) from Avian species. Toxicol. Sci. 2007, 99 (1), 101−117. (13) Lee, J. S.; Kim, E. Y.; Iwata, H. Dioxin activation of CYP1A5 promoter/enhancer regions from two avian species, common cormorant (Phalacrocorax carbo) and chicken (Gallus gallus): Association with aryl hydrocarbon receptor 1 and 2 isoforms. Toxicol. Appl. Pharmacol. 2009, 234 (1), 1−13. (14) Lee, J. S.; Iwabuchi, K.; Nomaru, K.; Nagahama, N.; Kim, E. Y.; Iwata, H. Molecular and functional characterization of a novel aryl hydrocarbon receptor isoform, AHR1β, in the chicken (Gallus gallus). Toxicol. Sci. 2013, 136 (2), 450−66. (15) Van den Berg, M.; Birnbaum, L.; Bosveld, A. T.; Brunström, B.; Cook, P.; Feeley, M.; Giesy, J. P.; Hanberg, A.; Hasegawa, R.; Kennedy, S. W.; Kubiak, T.; Larsen, J. C.; Van Leeuwen, F. X.; Liem, A. K.; Nolt, C.; Peterson, R. E.; Poellinger, L.; Safe, S.; Schrenk, D.; Tillitt, D.; Tysklind, M.; Younes, M.; Waern, F.; Zacharewski, T. Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environ. Health Perspect. 1998, 106 (12), 775−792. (16) Head, J. A.; Kennedy, S. W. Correlation between an in vitro and an in vivo measure of dioxin sensitivity in birds. Ecotoxicology 2010, 19 (2), 377−382. (17) Lee, J. S.; Kim, E. Y.; Iwabuchi, K.; Iwata, H. Molecular and functional characterization of aryl hydrocarbon receptor nuclear translocator 1 (ARNT1) and ARNT2 in chicken (Gallus gallus). Comp. Biochem. Physiol., Part C: Toxicol. Pharmacol. 2011, 153 (3), 269− 279. (18) Goto, J.; Kataoka, R.; Muta, H.; Hirayama, N. ASEDock-docking based on alpha spheres and excluded volumes. J. Chem. Inf. Model. 2008, 48 (3), 583−590. (19) Labute, P. LowModeMDimplicit low-mode velocity filtering applied to conformational search of macrocycles and protein loops. J. Chem. Inf. Model. 2010, 50 (5), 792−800. (20) Beglov, D.; Roux, B. An Integral Equation to Describe the Solvation of Polar Molecules in Liquid Water. J. Phys. Chem. B 1997, 101 (39), 7821−7826. (21) Kovalenko, A.; Hirata, F. Three-dimensional density profiles of water in contact with a solute of arbitrary shape: A RISM approach. Chem. Phys. Lett. 1998, 290 (1−3), 237−244. (22) Kovalenko, A.; Hirata, F. Self-consistent description of a metalwater interface by the Kohn-Sham density functional theory and the three-dimensional reference interaction site model. J. Chem. Phys. 1999, 110, 10095−10112. (23) Luchko, T.; Gusarov, S.; Roe, D. R.; Simmerling, C.; Case, D. A.; Tuszynski, J.; Kovalenko, A. Three-dimensional molecular theory of solvation coupled with molecular dynamics in Amber. J. Chem. Theory Comput. 2010, 6 (3), 607−624. (24) Bisson, W. H.; Koch, D. C.; O’Donnell, E. F.; Khalil, S. M.; Kerkvliet, N. I.; Tanguay, R. L.; Abagyan, R.; Kolluri, S. K. Modeling of the aryl hydrocarbon receptor (AhR) ligand binding domain and its utility in virtual ligand screening to predict new AhR ligands. J. Med. Chem. 2009, 52 (18), 5635−5641. (25) Pandini, A.; Soshilov, A. A.; Song, Y.; Zhao, J.; Bonati, L.; Denison, M. S. Detection of the TCDD binding fingerprint within the Ah receptor ligand binding domain by structurally driven mutagenesis and functional analysis. Biochemistry 2009, 48 (25), 5972−5983. (26) Motto, I.; Bordogna, A.; Soshilov, A. A.; Denison, M. S.; Bonati, L. A new aryl hydrocarbon receptor homology model targeted to improve docking reliability. J. Chem. Inf. Model. 2011, 51 (11), 2868−2881. (27) Xing, Y.; Nukaya, M.; Satyshur, K. A.; Jiang, L.; Stanevich, V.; Korkmaz, E. N.; Burdette, L.; Kennedy, G. D.; Cui, Q.; Bradfield, C. A.

of TCDD in avian AHRs and mouse AHR (Figure S8); binding modes of FICZ in avian AHRs and mouse AHR (Figure S9); transactivation potencies of ckAHR1 and ccAHR1 by TCDD and FICZ measured by in vitro reporter gene assays (Figure S10); relationship between potential TCDD−AHR interaction energies (Eprot-lig) and AHR-mediated transactivation potencies of TCDD (EC50) (Figure S11); predicted distributions of water molecules in AHR LBDs (Figure S12). This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +822-961-2310; fax: +822-961-0244; e-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Funding support was provided to E.-Y.K. from the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2013R1A1A2A10010043 and 2012K2A2A4021504). This work was also supported by Grants-in-Aid for Scientific Research (S) (21221004 and 26220103) and Challenging Exploratory Research (25660228) to H.I. and Postdoctoral Fellowships to M.H. from Japan Society for the Promotion of Science (JSPS). Financial assistance was also provided from Joint Research Project under the Japan-Korea Basic Scientific Cooperation Program for FY 2012 from JSPS and NRF. The first author (M.H.) is a JSPS research fellow.

■

REFERENCES

(1) Kubota, A.; Iwata, H.; Tanabe, S.; Yoneda, K.; Tobata, S. Levels and toxicokinetic behaviors of PCDD, PCDF, and coplanar PCB congeners in common cormorants from Lake Biwa, Japan. Environ. Sci. Technol. 2004, 38 (14), 3853−3859. (2) Kunisue, T.; Nakanishi, S.; Oka, N.; Sato, F.; Tsurumi, M.; Tanabe, S. Dioxins and related compounds in albatrosses from the Torishima Island, Japan: Accumulation features by growth stage and toxicological implications. Environ. Sci. Technol. 2006, 40 (22), 6919−6927. (3) Giesy, J. P.; Ludwig, J. P.; Tillitt, D. E. Deformities in birds of the Great Lakes region. Assessing causality. Environ. Sci. Technol. 1994, 28 (3), 128A−135A. (4) Leena Mol, T.; Kim, E. Y.; Ishibashi, H.; Iwata, H. In vitro transactivation potencies of black-footed albatross (Phoebastria nigripes) AHR1 and AHR2 by dioxins to predict CYP1A expression in the wild population. Environ. Sci. Technol. 2012, 46 (1), 525−533. (5) Thuruthippallil, L. M.; Kubota, A.; Kim, E. Y.; Iwata, H. Alternative in vitro approach for assessing AHR-mediated CYP1A induction by dioxins in wild cormorant (Pharacrocorax carbo) population. Environ. Sci. Technol. 2013, 47 (12), 6656−6663. (6) Furness, S. G. B.; Lees, M. J.; Whitelaw, M. L. The dioxin (aryl hydrocarbon) receptor as a model for adaptive responses of bHLH/PAS transcription factors. FEBS Lett. 2007, 581 (19), 3616−3625. (7) Karchner, S. I.; Franks, D. G.; Powell, W. H.; Hahn, M. E. Regulatory interactions among three members of the vertebrate aryl hydrocarbon receptor family: AHR repressor, AHR1, and AHR2. J. Biol. Chem. 2002, 277 (9), 6949−6959. (8) Hankinson, O. The aryl hydrocarbon receptor complex. Annu. Rev. Pharmacol. Toxicol. 1995, 35, 307−340. (9) Denison, M. S.; Pandini, A.; Nagy, S. R.; Baldwin, E. P.; Bonati, L. Ligand binding and activation of the Ah receptor. Chem.-Biol. Interact. 2002, 141 (1−2), 3−24. 3803

DOI: 10.1021/es505733f Environ. Sci. Technol. 2015, 49, 3795−3804

Article

Environmental Science & Technology

domain residues confer low dioxin responsiveness to AHR1β of Xenopus laevis. Biochemistry 2013, 52 (10), 1746−1754. (41) Kim, E. Y.; Iwata, H.; Yasui, T.; Inoue, N.; Lee, J. S.; Franks, D. G.; Karchner, S. I.; Hahn,M. E.; Tanabe, S. Molecular basis for differential dioxin sensitivity in birds: Characterization of avian AHR isoforms. In Interdisciplinary Studies on Environmental Chemistry: Vol. 1 Biological Responses to Chemical Pollutants; Murakami, Y., Nakayama, K., Kitamura, S.-I., Iwata, H., Tanabe, S., Eds.; Terrapub: Tokyo, 2008; Vol. 1, pp 81− 86. (42) Sindhikara, D. J.; Hirata, F. Analysis of biomolecular solvation sites by 3D-RISM theory. J. Phys. Chem. B 2013, 117 (22), 6718−6723.

Identification of the Ah-receptor structural determinants for ligand preferences. Toxicol. Sci. 2012, 129 (1), 86−97. (28) Fraccalvieri, D.; Soshilov, A. A.; Karchner, S. I.; Franks, D. G.; Pandini, A.; Bonati, L.; Hahn, M. E.; Denison, M. S. Comparative analysis of homology models of the Ah receptor ligand binding domain: Verification of structure−function predictions by site-directed mutagenesis of a nonfunctional receptor. Biochemistry 2013, 52 (4), 714− 725. (29) Farmahin, R.; Manning, G. E.; Crump, D.; Wu, D.; Mundy, L. J.; Jones, S. P.; Hahn, M. E.; Karchner, S. I.; Giesy, J. P.; Bursian, S. J.; Zwiernik, M. J.; Fredricks, T. B.; Kennedy, S. W. Amino acid sequence of the ligand-binding domain of the aryl hydrocarbon receptor 1 predicts sensitivity of wild birds to effects of dioxin-like compounds. Toxicol. Sci. 2013, 131 (1), 139−152. (30) Bessede, A.; Gargaro, M.; Pallotta, M. T.; Matino, D.; Servillo, G.; Brunacci, C.; Bicciato, S.; Mazza, E. M. C.; Macchiarulo, A.; Vacca, C.; Iannitti, R.; Tissi, L.; Volpi, C.; Belladonna, M. L.; Orabona, C.; Bianchi, R.; Lanz, T. V.; Platten, M.; Fazia, M. A. D.; Piobbico, D.; Zelante, T.; Funakoshi, H.; Nakamura, T.; Gilot, D.; Denison, M. S.; Guillemin, G. J.; DuHadaway, J. B.; Prendergast, G. C.; Metz, R.; Geffard, M.; Boon, L.; Pirro, M.; Iorio, A.; Veyret, B.; Romani, L.; Grohmann, U.; Fallarino, F.; Puccetti, P. Aryl hydrocarbon receptor control of a disease tolerance defence pathway. Nature 2014, 511, 184−190. (31) Moura-Alves, P.; Faé, K.; Houthuys, E.; Dorhoi, A.; Kreuchwig, A.; Furkert, J.; Barison, N.; Diehl, A.; Munder, A.; Constant, P.; Skrahina, T.; Guhlich-Bornhof, U.; Klemm, M.; Koehler, A.-B.; Bandermann, S.; Goosmann, C.; Mollenkopf, H.-J.; Hurwitz, R.; Brinkmann, V.; Fillatreau, S.; Daffe, M.; Tümmler, B.; Kolbe, M.; Oschkinat, H.; Krause, G.; Kaufmann, S. H. E. AhR sensing of bacterial pigments regulates antibacterial defence. Nature 2014, 512, 387−392. (32) Karchner, S. I.; Franks, D. G.; Kennedy, S. W.; Hahn, M. E. The molecular basis for differential dioxin sensitivity in birds: Role of the aryl hydrocarbon receptor. Proc. Natl. Acad. Sci. U.S.A. 2006, 103 (16), 6252−6257. (33) Van den Berg, M.; Birnbaum, L. S.; Denison, M.; De Vito, M.; Farland, W.; Feeley, M.; Fiedler, H.; Hakansson, H.; Hanberg, A.; Haws, L.; Rose, M.; Safe, S.; Schrenk, D.; Tohyama, C.; Tritscher, A.; Tuomisto, J.; Tysklind, M.; Walker, N.; Peterson, R. E. The 2005 World Health Organization Reevaluation of Human and Mammalian Toxic Equivalency Factors for Dioxins and Dioxin-Like Compounds. Toxicol. Sci. 2006, 93 (2), 223−241. (34) Ema, M.; Ohe, N.; Suzuki, M.; Mimura, J.; Sogawa, K.; Ikawa, S.; Fujii- Kuriyama, Y. Dioxin binding activities of polymorphic forms of mouse and human arylhydrocarbon receptors. J. Biol. Chem. 1994, 269 (44), 27337−27343. (35) Goryo, K.; Suzuki, A.; Del Carpio, C. A.; Siizaki, K.; Kuriyama, E.; Mikami, Y.; Kinoshita, K.; Yasumoto, K.; Rannug, A.; Miyamoto, A.; Fujii-Kuriyama, Y.; Sogawa, K. Identification of amino acid residues in the Ah receptor involved in ligand binding. Biochem. Biophys. Res. Commun. 2007, 354 (2), 396−402. (36) Pandini, A.; Denison, M. S.; Song, Y.; Soshilov, A. A.; Bonati, L. Structural and functional characterization of the aryl hydrocarbon receptor ligand binding domain by homology modeling and mutational analysis. Biochemistry 2007, 46 (3), 696−708. (37) Nuti, R.; Gargaro, M.; Matino, D.; Dolciami, D.; Grohmann, U.; Puccetti, P.; Fallarino, F.; Macchiarulo, A. Ligand binding and functional selectivity of L-tryptophan metabolites at the mouse aryl hydrocarbon receptor (mAhR). J. Chem. Inf. Model. 2014, 54, 3373−3383. (38) Farmahin, R.; Crump, D.; Kennedy, S. W. Sensitivity of avian species to the aryl hydrocarbon receptor ligand 6-formylindolo [3,2-b] carbazole (FICZ). Chem.-Biol. Interact. 2014, 221C, 61−69. (39) Perkins, A.; Philips, J. L.; Kerkvliet, N. I.; Tanguay, R. L.; Perdew, G. H.; Kulluri, S. K.; Bisson, W. H. A structural switch between agonist and antagonist bound conformations for a ligand-optimized model of the human aryl hydrocarbon receptor ligand binding domain. Biology 2014, 3, 645−669. (40) Odio, C.; Holzman, S. A.; Denison, M. S.; Fraccalvieri, D.; Bonati, L.; Franks, D. G.; Hahn, M. E.; Powell, W. H. Specific ligand binding 3804

DOI: 10.1021/es505733f Environ. Sci. Technol. 2015, 49, 3795−3804