Modeling, molecular dynamics and docking studies of a full-length Echinococcus granulosus 2DBD nuclear receptor

Abstract Nuclear receptors are ligand-activated transcription factors capable of regulating the expression of complex gene networks. The family includes seven subfamilies of protein with a wide phylogenetic distribution. A novel subfamily with two DNA-binding domains (2DBDs) has been first reported in Schistosoma mansoni (Platyhelminth, Trematoda). Employing an ab initio protocol and homology modeling methods, the full-length 3D structure of the Eg2DBDα.1 nuclear receptor from Echinococcus granulosus (Platyhelminth, Cestoda) was generated. The model analysis reveals the presence of the conserved three-layered alpha-helical sandwich structure in the ligand binding domain, and a particularly long and flexible hinge region. Molecular dynamics simulations were performed previous to dock a conformational library of fatty acids and retinoic acids. Our results indicate that oleic and linoleic acids are suitable ligands to this receptor. The ligand-protein complex is stabilized mainly by hydrogen bonds and hydrophobic interactions. The fact that 2DBD nuclear receptors have not been identified in vertebrates confers particular interest to these nuclear receptors, not only concerning their structure and function but as targets of new anthelmintic drugs. Communicated by Ramaswamy H. Sarma


Introduction
More than 900 nuclear receptors (NRs) genes have been identified in all animals examined (Mazaira et al., 2018;Sladek, 2011). The number of NRs expressed by different species of vertebrates and invertebrates varies considerably; humans have 48 members while Drosophila 18 and Caenorhabditis 270 (Brehm & Koziol, 2017;King-Jones & Thummel, 2005;Robinson-Rechavi et al., 2003;Sluder et al., 1999). The family includes seven subfamilies of proteins with a wide phylogenetic distribution. The most prominent distinction from other transcription factors is their capacity to specifically bind small lipophilic ligands such as steroids, retinoids, phospholipids, fatty acids, cholesterol, etc. However this family of proteins also contains orphan receptors for which ligands have not yet been identified (Evans & Mangelsdorf, 2014;Weikum et al., 2018). Nuclear receptors are of greatest importance for metazoan intercellular signaling as they initiate and regulate diverse aspects of a multicellular organism life, such as metabolism, homeostasis, cell proliferation, immune response, enzyme activity, development, and reproduction, as well as many pathological processes (Mazaira et al., 2018). There have been multiple structural studies of nuclear receptors involving either ligand binding domain (LBD) and DNA binding domain (DBD) fragments alone, but few fulllength structures (Chandra et al., 2008(Chandra et al., , 2013Lou et al., 2014). NRs share a common modular organization. A poorly structured N-terminal domain (region A/B) contains the ligand-independent activator function-1 region (AF-1), which interacts with co-regulator proteins in a cell and promoter specific manner (Kumar & Thompson, 2003;Weikum et al., 2018). This domain is followed by the DNA binding domain (region C) composed of two zinc fingers which is the hallmark of the nuclear receptor family since it is the most conserved among all nuclear receptor domains. The DBD contains a P-box, a short motif responsible for direct DNA interaction and DNA-binding specificity, and a D-box implicated in dimerization pattern. A flexible hinge region (region D) links the DNA binding domain to the ligand binding domain (region E). LBD has a general folding consisting of a three-layered antiparallel alpha-helical sandwich composed of 12 alpha-helices, including a beta-sheet (2 beta-strands) (Pawlak et al., 2012). The LBD also contains a strong dimerization interface and a variable ligand binding pocket (LBP) that allows NRs to recognize different ligands (Wurtz et al., 1996). Some nuclear receptors contain a poorly known function region (F region) in the C-terminal (Germain et al., 2006).
In 2006, a novel subfamily of nuclear receptor named 2DBD with members containing two tandem DNA-binding domains (2DBDs) was reported in Schistosoma mansoni (Platyhelminth, Trematoda) (Wu et al., 2006). Subsequently, members of this subfamily have been identified in some mollusks, arthropods, and platyhelmithes (Alvite et al., 2019;Wu et al., 2007). Interestingly, these atypical NRs are not present in vertebrates. This new subfamily of NRs has not yet been included in the nuclear receptor classification system (Nuclear Receptor Nomenclature Committee, 1999). These proteins possess the typical architecture and conserved motifs of the NRs. Four 2DBD receptors were identified in the parasitic platyhelminth Echinococcus granulosus (Eg2DBDa, Eg2DBDa.1, Eg2DBDb and Eg2DBDc), being Eg2DBDa.1 an isoform of Eg2DBDa detected in the protoescolex larval stage (Alvite et al., 2019). E. granulosus is the causative agent of cystic echinococcosis, a cosmopolitan zoonosis that constitutes a major public health problem and a cause of significant economic loss (Busto Bea et al., 2016). To date, only two NRs have been characterized in this parasitic platyhelminth (F€ orster et al., 2011;Yang et al., 2017).
NRs have become attractive targets for the development of small molecular for oral use, due to the lipophilic nature of their ligands and their ability to modulate the expression of multiple genes in the same pathway (Evans & Mangelsdorf, 2014). Parasitic helminth NRs are interesting targets for anthelmintic drugs since new ones are required to treat these infections due to the lack of effective therapeutic strategies . Moreover, the presence of two DNA binding domains confers particular interest to these nuclear receptors, not only concerning their unknown function but to the development of new anthelmintic drugs. All 2DBD-NRs have the same P-box sequence in the first DBD that is not present in another known NR (Wu et al., 2006;2007) suggesting that this characteristic P-box could determine a new target DNA binding specificity (Wu & LoVerde, 2011).
In this work, we describe the three-dimensional modelled structure of the recently reported Eg2DBDa.1 nuclear receptor from the parasite E. granulosus. The conformational structural map has been explored by molecular dynamics (MD) simulations. Furthermore, molecular docking studies were performed to assess putative ligands. The reported structure is the first full-length 3D model of a NR belonging to the 2DBD subfamily.

3D model structure
The amino acid sequence of Eg2DBDa.1 (GenBank accession number AZM65758.2) was employed to search structurally homologous sequences in the Protein Data Bank using the Sequences Annotated by Structure server available at EBI (https://www.ebi.ac.uk/). The amino acid similarity between the target and the best templates was observed to be less than 40%. For this reason, and in order to accomplish a reliable 3D model of this novel NR we used a mixed strategy by threading and homology modelling. The Iterative Threading Assembly Refinement (I-TASSER) server was used (Yang & Zhang, 2015;Zhang et al., 2017). This software applies an Ab initio protocol for modeling 3D structures based on algorithms that predict structure and function of proteins.
Homology DBD modeled structure was obtained as previously described (Alvite et al., 2019), including that of DNA double strand. The SAS analysis for DBD I and DBD II domains, gives the alignment of 20 PDB structures with identity percentages ranging from 45.1-51.4 and 49.3-53.7%, respectively. With respect to LBD, since the percentage of identity with our sequence is under 25%, we used I-TASSER server to model it. The results bring retinoic acid receptor beta (PDB ID: 1XDK) as the template hit for modeling the LBD. Ab initio LBD model was obtained by I-TASSER server with the following parameters (LBD: TM-score ¼ 0.91, C score ¼ 0.97, RMSD ¼ 1.87 Å) and refined by means of an energy minimization in MOE ChemComp suite package (molecular operating environment (MOE), Chemical Computing Group Inc., Labute, 2014), AMBER12:EHT force field, gradient conjugate algorithm with a convergence gradient of 0.01 kcal/mol.
To obtain the full-length receptor model three steps were performed: a) an ab initio protocol from the I-TASSER server for the full-length protein (830 amino acids) was employed; b) the N-terminal region, Hinge and C-terminal region were taken from the ab initio full-length structure; c) DBD I, DBD II, and LBD regions obtained as mentioned above were remodeled over the corresponding sequences of the full model. The N-terminal, DBDI, DBDII, Hinge, LBD and C-terminal regions were defined in the complete three dimensional structure. PROCHECK was then used to analyze the structural and stereochemical properties by recognizing overall and residue-by-residue geometry. A Psi/Phi Ramachandran plot was used to assess the local quality of the model (supporting material Figure S1).The reliability of the model was assessed via ERRAT, which examines the statistics of nonbonded interfaces between diverse atom types. The plot shows that only 5.5% of the total residues falls under disallowed regions while most of the residues falls under most favored regions 60.7% and the other 33.8% of residues in additional and generously allowed regions

Molecular dynamics simulations
Complete simulation box was built with QwikMD from VMD (Humphrey et al., 1996). The full length model was solvated with TIP3P water molecules within octahedral boxes such that the solute molecules were at least 20 Å from the box boundaries. Sodium and chloride ions at 0.15 mM were added in order to make the systems electrically neutral at physiological conditions. The parameter 'rigidBonds' was used in 'on' to constrain the hydrogen bonding of nonpolar hydrogens. In the case of water, they were considered as rigid bodies molecules.
The complete system consisted of approximately 146.424 atoms (full-length complex).
Simulations were performed with NAMD (Phillips et al., 2005) using a 2 femtosecond (fs) time step for the integration of the equations of motion. CHARMM parameters were applied for protein and DNA (Brooks et al., 1983). The particle-mesh Ewald (PME) method was used to calculate the long-range electrostatic interactions beyond a cutoff of 12 Å. Periodic boundary conditions were applied with an isothermal-isobaric (NPT) ensemble at 300 K and 1 atm pressure maintained using the Langevin thermostat and Langevin Piston barostat, respectively (Ryckaert et al., 1977).
Minimization and equilibration procedures consisted of: i) 1000 gradient conjugate minimization steps; ii) 600 steps of annealing; iii) 500 ps of equilibration of the whole box with harmonic restraints in the backbone; iv) 200 nanoseconds (ns) of production without any restriction. After these procedures, 3 independent production trajectories were generated using different random seeds for the initial velocities to get good sampling of coordinate space lasting 200 ns each for a total simulation time of 0.6 microseconds. Post processing and analysis were carried out using cpptraj module from AmberTools14 and VMD 1.9.4 package used for trajectory visualization. Pictures emerging from the molecular dynamics as well as those from docking were processed with VMD 1.9.4 and MOE packages. Energetic (total and potential energy) and structural analyses were performed to assess the stability of the system: graphical inspection, root mean square deviations (RMSD), root mean square fluctuations (RMSF) and hydrogen bonds.
After molecular dynamics (MD) and analysis, a full sequence structure was obtained and used to further Site Finder, Molecular docking and Protein Ligand Interaction Fingerprints (PLIF) studies.

Ligand binding site prediction
The Site Finder option on MOE was used to calculate possible binding sites by generating the alpha-site spheres in the 3D ligand binding domain structure from the atomic coordinates of Eg2DBDa.1. The following settings were used: two probe radii of 1.4 and 1.8 Å, a distance to isolate donor and acceptor features of 3 Å, a connection distance of 2.5 Å and a minimum site size of 30 alpha sphere with a 2 Å radius. All sites whose maximum distance to the mean site point is less than 2 Å were removed.

Molecular docking
Interested in the native putative ligands of the nuclear receptor studied, several fatty acids were selected to perform molecular docking. The selected ligands were identified in E. granulosus larval stage and also are natural ligands of nuclear receptors (Thompson, 1995;Weikum et al., 2018). The following molecules were assayed: Oleic Acid (OA), Palmitic Acid (PA), Arachidonic Acid (AA), Linoleic Acid (LA) and 9-cis and 13-cis-Retinoic Acid (9-RA and 13-RA). A database was generated using drawing tools from MOE. The ligands were charged, protonated and energy minimized through MOE, using the Amber12-EHT force field and other default parameters. Each molecule was submitted to a conformational search in which all rotatable angles are assayed in the space using the LowModeMD conformational search, a conformational search method that uses implicit vibrational analysis to focus a MD trajectory along the low-mode vibrations. This procedure has the effect of searching for minima along the valleys and troughs on the potential energy surface. The forcefield partial charges were calculated, and a set of limits (Rejection Limit ¼ 100, Iteration Limit ¼ 10,000, RMS Gradient ¼ 0.05 and MM Iteration Limit of 500) and other definitions (RMSD ¼ 0.25 Å, a limit to judge two conformations as equal, a window cutoff of 7 (E_min þ cutoff) to discard conformations with greater energies than E_Min þ cutoff (where E_min is the global minimum)) were set.
The 3D modeled structure of the full-length Eg2DBDa.1 nuclear receptor was protonated and the energy minimization was performed using AMBER12:EHT force field and default parameters from MOE software. After, a molecular dynamics as above described was used as a validation process of this structure, and the structure from 100 ns trajectory was selected as the receptor structure for docking. A Site Finder procedure was performed over this structure and 16 sites were obtained. After revision, the Site 7 was selected for docking. The Alpha PMI function was selected for placement. London dG was used to score each placed ligand conformation and the GBVI/WSA dG function was used to rescore. After docking, energy minimization and removal of duplicate conformations, only 10,000 lowest energy conformations were retained. The resulting docked database for all ligands was obtained and it was annotated by the receptorligand docking score. This database was used to study the ligand conformations and their final score, the 2D and 3D ligand binding interactions were saved for analysis and to obtain the Protein Ligand Interaction Fingerprints.

Protein ligand interaction fingerprints
The Protein Ligand Interaction Fingerprints (PLIF) tool allows summarizing the interactions between ligands and proteins using a fingerprint scheme. PLIF analysis was performed by MOE package. All docked conformations with Score interaction energies between -10 and -7.6 kcal/mol were used as the input data for PLIF. Hydrogen bonds between polar atoms are calculated using a method based on protein contact statistics, whereby a pair of atoms is scored by distance and orientation. The score is expressed as a percentage probability of being a good hydrogen bond (Labute, 2001). Ionic interactions are scored by calculating the inverse square of the distance between atoms with opposite formal charge and expressing as a percentage (100% corresponds to 1 Å distance). Surface contact interactions are determined by calculating the solvent exposed surface area of the residue, first in the absence of the ligand, then in the presence of the ligand. The difference between the two values is the extent to which the ligand has shielded the residue from exposure to solvent, which is potentially indicative of a hydrophobic interaction. The solvent exposed surface area is determined by adding 1.4 Å to the van der Waals radii of each heavy atom, and computing the fraction of this total surface which does not lie within the radius of any other.

Energetic and structural analysis of simulations
The time course evolution of the total energy along the trajectory of three independent simulations suggests that the full-length protein-DNA complex structure achieved thermal equilibration immediately after the equilibration phase ended demonstrating that the simulation was physically valid (Figure 1(A)).
Average values for the complexes were about -1.5 Â 10 4 kcal/mol for each replicate. In all cases, the energy curves for the different trajectories of the system are practically superimposed. This is consistent with the fact that all trajectories obtained for the nuclear receptor model trajectories are sampling the same thermodynamic state and, in particular, the explored regions in the structural landscape show very similar energy levels. The full-length protein-DNA complex structure achieves thermal equilibration after 20 ns of MD simulation remaining stable throughout the course of the trajectories.
Graphical inspection and backbone RMSD revealed that the full-length protein-DNA complex structure achieved equilibration after 50 ns of MD simulation remaining stable throughout the course of the trajectories. The RMSD of full trajectory taken as reference the energy minimized structure was of 9.11 A ± 2.09 Å and when the averaged structure was taken as reference the result was of 4.66 ± 1.10 Å. As it could be observed in Figure 1(B) these parameters remain constant from about 50 ns trajectory. Moreover, a partial RMSD analysis for the DBD and LBD domains was performed. Taking the contact DNA region within 16 Å from the nucleic acid and re-evaluating de RMSD for this specific region, the result was 5.01 ± 0.6 Å and 2.65 ± 0.57 Å with the energy minimized or the averaged structure as reference, respectively. For the LBD domain the same procedure resulted in 4.12 ± 0.65 Å or 1.94 ± 0.48 Å.

Inter-domains structural dynamics
Molecular dynamics simulations of full-length Eg2DBDa.1 structure bound to putative DNA response-elements provided us information about the conformational disposition and organization of each domain. This is highly valuable since there is a small number of available full-length NR crystal structures due to the challenge it represents the crystal packing of A/B and hinge regions, known to be quiet disordered and disturb.
In order to get close to the mobility and dynamic of the complex, overall RMSF values were calculated for the backbone and lateral chain atoms of each residue except hydrogens. Higher RMSF values indicate greater flexibility during the MD simulation ( Figure 2). RMSF analysis shows that the N-terminal region (A/B) as well as the Hinge (D) and C-terminal (F) regions are highly dynamic suggesting that they undergo large conformational changes regarding DBDs and LBD regions. These results are in concordance with reported full-length crystallographic structures (Chandra et al., 2008(Chandra et al., , 2013Lou et al., 2014).

Structural analysis of Eg2DBDa.1 model
Despite the existence of several papers concerning structural studies of nuclear receptors LBD and DBD regions, few fulllength NR structures have been reported (Chandra et al., 2008; 2013; Lou et al., 2014). Figure 3 depicts the full-length Eg2DBDa.1 model indicating the corresponding sequence, and a magnified view of the LBD.
The validated model proposes a three-dimensional putative structure for the native protein. The structure contains the main characteristics of the nuclear receptor family (Mazaira et al., 2018). A long N-terminal domain (170 residues) folds in several alpha-helical regions. DBD domain conforms to four Zn fingers previously described (Alvite et al., 2019). Briefly, the first alpha-helix of DBD I engage the major groove side of the DNA, making base-specific interactions between Lysines 194-195 and two Guanines from the consensus response element. In addition, DBD II-DNA interaction could contribute to stabilize the complex. Few reports describe the Hinge region of nuclear receptors. This poorly conserved domain serves as a hinge between the DBD and the LBD, allowing the adoption of different relative positions to avoid steric hindrance troubles (Germain et al., 2006). In PPAR-c, the hinge region forms a significant DNA interaction followed by two helical segments that reach the LBD, while RXR-a hinge is devoid of secondary structure and is more flexible (Chandra et al., 2008). Eg2DBDa.1 hinge fold resembles that of RXR-a receptor probably conferring great flexibility. This region of 132 residues is longer in length than most known higher organism nuclear receptors.
The LBD domain structure has been widely studied (Moras & Gronemeyer, 1998;Weatherman et al., 1999;Weikum et al., 2018;Wurtz et al., 1996). It is a complex structure that not only binds to ligands but also interacts directly with co-regulator proteins. This structurally conserved domain contains a general folding structured by 2 betastrands and 12 alpha-helices (Pawlak et al., 2012). However, Eg2DBDa.1-LBD structure presents 2 beta-strands and 11 alpha-helices with similar canonical folding (Figure 3(B)). In order to maintain the consensus helices numbers, we omit helix (H) 11 referring to it as H12. As was described for multiple other NR LBD solved structures, the helical sandwich bundle is conserved, where Eg2DBDa.1-LBD H1 and H3 form one face, s1-s2, H5, H6 and H8 are in the central layer of the domain and H7 and H10 constitute the second face. Previous alignment with parasites and vertebrates nuclear receptor sequences suggested that LBD H1 should begin on Gly489 (Alvite et al., 2019). However, once assembled the full-length 3D structure, this helix begins on Ala455. It is very striking the fact that this helix is only composed by a track of Alanines. The AF-2 conserved motif identified in this receptor as the sequence LYVEMY is localized in the helix 12. It is worth mentioning that H12 has high mobility according to RMSF analysis suggesting that it could fulfill the activation role as it does in other nuclear receptors (Fischer & Smie sko, 2020;Moras & Gronemeyer, 1998). The AF-2 site corresponds to a protein-protein interaction surface for the binding of coactivator proteins essential for downstream signaling, which renders it an attractive target for potential inhibitors (Fisher & Smie sko, 2020). Eg2DBDa.1 F region (105 amino acids) is highly dynamic suggesting that it could undergo conformational changes being able to influence the receptor's activity by affecting transcriptional activation, dimerization, interactions with other proteins and/or the stabilization of agonist or antagonist bound conformations in the receptor LBD, as was reported for other NRs (Patel & Skafar, 2015). Interestingly, the available information about the F domains of NRs indicates that this region presents different lengths but a common function as regulator of transcriptional activity.

Docking of fatty acids and retinoic acids
Site Finder procedure detected 27 sites in the receptor. Most of them (22/27) and the ones with larger volume and surface areas belong to the LBD domain of Eg2DBDa.1 modeled structure (Figure 4(A)) suggesting that this domain has the highest capacity to bind putative ligands. The site number seven (Site7) (Figure 4(B)) was selected as 'Site' for docking studies since it was the most similar to several reported ligand binding pockets in several solved structures (1upv; 2b50; 1xdk; 6nwk; 3gwx).
For the docking process, a conformational database of fatty acids and retinoic acids (37499 conformations) was used. For each ligand, out of many docking poses, only those that possessed the highest docking score were chosen. Among the ligands under study, the OA showed the highest binding affinity to Eg2DBDa.1 with a score of -10.0947 kcal/ mol, followed by LA (-9.6666 kcal/mol), AA (-9.5347 kcal/mol) and PA (-8.0873 kcal/mol) in the fourth place. Retinoic acids showed scores lower than -6.65 kcal/mol demonstrating that fatty acids are better ligands for this cestode nuclear receptor. Polar head of OA and LA interacts with Tyr566 from H5 and Phe508 from the loop H2-H3. The fatty acid tail is surrounded by many H3 polar residues and no polar residues from H3, H6 and connecting loops (Figures 5 and 6). LA and AA adopt a similar position facing the polar head to H3. LA links with Tyr566 and His531. The other assayed ligands adopted an opposite orientation, facing to loops H1-H2 and H2-H3. PA interacts with Arg507, Phe508, Gly489 and Met490. Figure 5 shows a two-dimensional diagram of docked fatty acid while Figure 6 represents the three-dimensional binding mode of oleic and linoleic acids.
Curiously, fatty acids studied exhibit two binding modes. OA, LA and ARA share a common orientation inside the binding pocket. The polar head of OA and LA ( Figure 5(A,B)) are making a hydrogen bond with Tyr566 from H5, while the hydrophobic tail is sustained by a cluster of aromatic and hydrophobic amino acids from H3, H6, and H7 as indicated above. His531 and His535 seem to sustain the hydrophobic body with their aromatic side chain. The other assayed ligands, PA and RAs, adopted an opposite orientation, pointing the carboxylate head towards Arg491 and Arg507 from loops H1-H2 and H2-H3.

Fingerprint analysis of fatty acids
Taking into account that the docking scoring algorithm may not be too sensitive, differences in binding affinities could remain underestimated. To have a more statistical evaluation of docked domains and to visualize the main set of conformations making the more energetic contacts, fingerprint analysis was also performed. Docking poses with binding affinities between -10 and -7.6 kcal/mol (52 poses) were selected. Interactions such as hydrogen bonds, ionic interactions and surface contacts are classified according to the residue of origin, and built into a fingerprint scheme which is representative of the given protein-ligand docked complexes. There are six types of interactions in which a residue may participate: side chain hydrogen bonds (donor or acceptor), backbone hydrogen bonds (donor or acceptor), ionic interactions, and surface interactions. The most potent interactions in each category, if any, are considered. If no interactions of a particular category are found, or none pass the thresholds, no bits are set for that category. If the strongest interaction passes the lower interaction threshold, the low order fingerprint bit is set. If the strongest interaction passes the higher interaction threshold, then the low order and high order bits are both set. Therefore the bit patterns for each category can take on values of 00, 10 or 11, respectively. The calculations of protein ligand interaction fingerprints highlight those amino acids from the ligand binding pocket that are important for the binding interactions (Figure 7). It is worth noticing that amino acids involved in the interaction of best docked fatty acid conformations are included. Particularly, residues Met492, Phe532, Phe586, Val597 and His600 were detected on site 7, docked binding sphere and PLIF analysis.
From the PLIF analysis it is very interesting to annotate the role of the aromatic ring of His531 making the 100% of interactions with the full set of conformations. Phe508, Phe532, Tyr566 and Phe586 are present here again making a crucial hydrophobic and aromatic interaction. Arg597, conforming the second binding mode, is forming hydrogen bonds with some conformations. Figure 7(B) graphically depicts these results emphasizing once again the role of Tyr566, His531 and Phe508 in the definition of the interaction. Physicochemical properties of these residues are in agreement with the amphipathic characteristic of fatty acids that are the putative ligands for the studied protein.

Conclusions
The present study shows the full-length structure of a member of the new 2DBD nuclear receptor subfamily employing an in silico approach. Docking studies suggested that unsaturated fatty acids are the preferred ligands by Eg2DBDa.1 Echinococcus granulosus nuclear receptor. This result is particularly interesting since previous studies also showed that OA, LA, and AA are fatty acid binding protein EgFABP1 preferred ligands (Alvite et al., 2001;Esteves et al., 2013). This protein could be responsible for translocating these fatty acids to the nucleus since it localizes in the nuclear subcellular fraction (Alvite & Esteves, 2016). In this sense, EgFABP1 could transfer the ligand to Eg2DBDa.1 activating the nuclear receptor. This fact also positions this receptor in a relevant place for the survival of the parasite, since it cannot synthesize fatty acids de novo. Taking into account that LBP region of nuclear receptors had been extensively studied as a target region of drugs against several diseases (Fisher & Smie sko, 2020), Eg2DBDa.1 emerged as a new target for specific drugs to fight cystic echinococcosis. It should be highlighted that 2DBD nuclear receptors have not been identified in vertebrates, taking force this hypothesis. The ligand protein interactions here reported could be used as a guideline for the design of Eg2DBDa.1 inhibitors, as well as a starting point to initiate functional studies to gain knowledge about the role of this particular nuclear receptor.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This work was supported by the National Agency for Research and Innovation (Uruguay) under Grant ANII-FCE-2017_1_136527.