• Users Online: 47
  • Print this page
  • Email this page


 
 Table of Contents  
ORIGINAL ARTICLE
Year : 2020  |  Volume : 7  |  Issue : 3  |  Page : 118-129

Computational Modeling of Trans-Zeatin as a Novel Target of Adenosine A2A Receptor: Insights into Molecular Interactions


1 Department of Bioengineering; Department of Molecular Biology and Genetics (English), Faculty of Engineering and Natural Sciences, Üsküdar University, Istanbul, Turkey
2 Department of Molecular Biology and Genetics (English), Faculty of Engineering and Natural Sciences, Üsküdar University, Istanbul, Turkey
3 Department of Chemical Engineering, Faculty of Engineering and Natural Sciences, Üsküdar University, Istanbul, Turkey

Date of Submission18-Aug-2020
Date of Decision13-Nov-2020
Date of Acceptance13-Nov-2020
Date of Web Publication25-Dec-2020

Correspondence Address:
Ebru Destan
Üsküdar University Central Campus, Altunizade Mah. Haluk Türksoy Sk. No: 14, Uskudar, İstanbul
Turkey
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/jnbs.jnbs_19_20

Rights and Permissions
  Abstract 


Background: Adenosine A2A receptor (A2AR) is a G-protein-coupled receptor that is involved in various physiological functions. Zeatin, a plant cytokinin and a derivative of adenine, is recently identified as new ligand of A2AR. However, the ligand-receptor interaction mechanism is not fully revealed. Aims and Objectives: The main aim of this study is to reveal a model structure of A2AR in complex with zeatin for the first time to provide a better understanding of this novel interaction mechanism. Materials and Methods: The model structure of A2AR in complex with zeatin was created by docking and the structural dynamics of the complex were detected by molecular dynamic simulations during the study. A model structure of A2AR in complex with caffeine was used as a positive control. Result: Zeatin displayed the ability to stay more stable at the binding pocket compared with caffeine based on molecular dynamic simulations and the residues involved in the interaction are identified, leading a new sight for further studies on zeatin and A2AR interaction. Conclusion: We propose that zeatin is indeed a novel and promising target for A2AR.

Keywords: Adenosine A2A receptor, binding pocket, caffeine, protein–ligand interaction, zeatin


How to cite this article:
Destan E, Öz P, Timuçin AC. Computational Modeling of Trans-Zeatin as a Novel Target of Adenosine A2A Receptor: Insights into Molecular Interactions. J Neurobehav Sci 2020;7:118-29

How to cite this URL:
Destan E, Öz P, Timuçin AC. Computational Modeling of Trans-Zeatin as a Novel Target of Adenosine A2A Receptor: Insights into Molecular Interactions. J Neurobehav Sci [serial online] 2020 [cited 2021 Apr 12];7:118-29. Available from: http://www.jnbsjournal.com/text.asp?2020/7/3/118/304919




  Introduction Top


Adenosine receptors (A1, A2A, A2B, and A3) are transmembrane G-protein-coupled receptors (GPCRs) that can either stimulate or inhibit adenylyl cyclase (AC) by their Gi and Gs subunits.[1] Adenosine A1 receptor and adenosine A3 receptor inhibit the AC via Gi subunit, whereas adenosine A2A receptor (A2AR) and adenosine A2B receptor stimulate the activity of AC via Gs subunit, which catalyze the production of cyclic adenosine monophosphate (cAMP).[2] A2ARs are expressed in higher density in the basal ganglia of the brain and in lower density in the cardiovascular and immune systems.[3],[4] A2AR consists of seven transmembrane domains with an extracellular amino terminus and a cytosolic carboxy terminus as a common feature with all GPCRs. The connection between transmembrane domains consists of between three extracellular and three cytoplasmic loops.[5] The binding of a ligand occurs on the extracellular side, leading to the conformational changes in the heptahelical transmembrane helix network of the receptor.[6]

Caffeine is a plant-derived methylxanthine and a well-known nonselective antagonist of A2AR.[7] A2AR has well-defined binding pocket and previous studies showed that caffeine interacts with the residues PHE 168, ILE 274, LEU 249, MET 270, ASN 253, TRP 246, and VAL 84 with an additional polar contact to HIS 278 in the hydrophobic pocket.[4],[8] Caffeine blocks the activity of A2AR, and therefore, it is considered an effective and widely consumed psychoactive drug and stimulant.[9] Caffeine can display neuroprotective effects by preventing the β-amyloid-induced neurotoxicity or promote wakefulness.[1],[10],[11],[12]

Zeatin, a plant phytohormone, promotes plant growth and development, was recently shown to interacts with A2AR; however, the binding mechanism has not been discovered yet.[13] Plant cytokinins are adenine derivatives substituted at the N6-position with either an isoprenoid or aromatic side chain, and cis- and trans-zeatin includes a substitution of the isoprenoid side chain.[13] Cytokinins have important antioxidative and protective effects in animals at molecular, cellular, tissue, and organismal levels.[14] Zeatin showed antioxidative and cell-protective effects against β-amyloid-induced neurotoxicity, similar to caffeine.[15] Most recently, the possible antidepressant effect of zeatin on female and male rats was shown, together with the interaction of zeatin with A2AR on the same binding site with caffeine.[16] These findings further suggest that zeatin exerts its effects via A2A-R-mediated downstream pathways. Zeatin can be converted to zeatin riboside by adenosine phosphorylase and zeatin riboside was shown to prevent the serum deprivation-induced apoptosis.[12] Previously, it has been known that A2ARs regulate CD4 + T lymphocyte, even suppressing the activation-induced cell death of peripheral T-cells.[17] In addition, zeatin riboside treatment promotes the production of cAMP in T-lymphocytes and inhibits the production in CD3 + CD4 + T-cells of interferons.[18] These findings clearly indicate an interaction between zeatin and A2AR, and a better understanding of this interaction can be achieved by structural modeling.


  Materials and Methods Top


Ethics committee approval

There is no need for ethics committee approval.

Structure preparation and docking

The crystal structure of A2AR (PDB entry: 5NLX) was used for docking of the ligands: trans-zeatin and caffeine. All other ligands found in the crystal structure of A2AR were removed to exert protein for the docking step via PyMol.[19] Trans-zeatin and caffeine were used as the ligand, while A2A was kept as the receptor. Structures of ligands were obtained from RCSB PDB database, is one of the largest protein databanks which offers to get PDB structures of proteins and their ligand, and ligand parameters were determined using SwissParam.[20] The crystal structure of A2AR that contains caffeine as a ligand (PDB entry: 5MZP) was used as a control during docking to detect possible binding pocket and connecting residues based on previously published alignments.[8],[21] The docking position of caffeine and trans-zeatin was detected in the same binding pocket and the docked structure of A2AR complex with caffeine was used as a positive control during the study. Docking was performed with AutoDock.[22] Final docking poses were selected based on AutoDock binding scores that give the best down binding energy and inhibition constant. The binding interactions that were performed during docking were analyzed by Arpeggio[23] and PyMOL.[19]

Designing of the membrane-bound protein

Membrane protein tutorial[24] was performed to docking structures of A2AR in complex with zeatin (5NLX/ZEA) and A2AR in complex with caffeine (5NLX/CFF), which were produced previously via AutoDock,[22] using Visual Molecular Dynamics (VMD).[25] To provide interaction between membrane and protein, the following steps were adapted from membrane protein tutorial:[24] generating PSF and PDB files ,are protein structure files which contains all structural data about the position of atom and residues, leading the detection of protein structure for molecular visualization systems, for building whole structure, building membrane patch for preparing complete membrane involving water around it, alignment of membrane and protein, a combination of membrane and protein for avoiding overlap between protein and lipid molecules, and solvation and ionization with 100 mM NaCI as the ionic concentration of the system. To perform molecular dynamic simulations, the structures were fixed according the membrane protein tutorial.[24]

Molecular dynamics simulations

Molecular dynamics (MD) simulations were performed with the structures that were obtained as a result of the previously described in the stage of designing of membrane-bound protein. 5NLX/ZEA and 5NLX/CFF complexes composed of 78,892 atoms and 78,757 atoms were placed in water boxes with dimensions of 83 Å ×94 Å ×120 Å and neutralized with NaCl. The resulting systems for 5NLX/ZEA and 5NLX/CFF were used in MD simulations using the NAMD[26] with the CHARMM22[27],[28] parameters included correction map for backbone atoms.[29],[30] TIP3P model[31] was used for water molecules within the system during the simulation. An NpT ensemble was used in MD simulations with periodic boundary conditions to maintain pressure and temperature, while the long-range Coulomb interactions were computed using the particle-mesh Ewald algorithm. Within 50 ns of MD simulation, the pressure was maintained at 1 atm and temperature was maintained at 310 K using the Langevin pressure and temperature coupling. The time step was determined as 1 fs in all MD simulations. For providing the removal of high-energy contacts between atoms and highly repulsive orientations of the initial simulated systems, the systems were fully energy minimized in 50,000,000 steps and each system was heated slowly from 0 K to 310 K in 5 ps. Then, the systems were equilibrated under constant temperature and volume for 0.5 ns before production runs. The production runs were completed for 50 ns as it is a large complex and repeated twice. In total, 300-ns MD simulation was performed during the study.

Detection of interacting residues with ligands: Zeatin and caffeine

The MD simulations which were repeated twice as previously described by using the NAMD[26] were analyzed to detect interacting residues between protein (5NLX) and ligands (zeatin and caffeine) by using VMD.[25] Hydrogen bonds, hydrophobic contacts, and salt bridges were analyzed in VMD[25] for the detection of interacting residues during the simulations. Then, these interactions were plotted using GraphPad Prism.[32] For the comparison of detected residues with published alignments, a known A2AR with caffeine (PDB entry: 5MZP) was used and the residues of 5MZP were renumbered as protein sequence was numbered as 1–305 in 5MZP, while protein sequence was numbered as 9–409 in our model.

Analysis of the molecular dynamics simulations

VMD[25] was used for the analysis of trajectories and the visualization of structures. Root mean square displacements (RMSDs) for the backbone atoms of each protein were analyzed for the stability in 50-ns MD simulations. Residue-specific distance between ligand (zeatin and caffeine) and protein was analyzed within 50 ns. In addition, the distance between the interacting atom of residue and the interacting atom of ligand (zeatin and caffeine) was analyzed. Residue-wise root mean square fluctuations (RMSFs) of ligands with protein were measured for the flexibility analysis of the ligand and t-test was applied to determine the significant difference between zeatin and caffeine values based on P < 0.05. The numerical data were expressed as mean ± standard error of mean in the graph. RMSFs of the detected residues of the protein were also analyzed within 20–50 ns based on the result of distance analysis because the time that ligands are unstable was detected by interpreting distance and RMSF was calculated within this range. Furthermore, radius of gyration was analyzed for both 5NLX/ZEA and 5NLX/CFF complexes. Distance, radius of gyration, and RMSF graphs were plotted by taking the average of three production runs for both 5NLX/ZEA and 5NLX/CFF complexes via GraphPad Prism,[31] while RMSD graphs were plotted for each production run respectively via GraphPad Prism.[32]


  Results Top


Trans-zeatin and caffeine were docked to the binding pocket of A2AR (PDB entry: 5NLX) as described [Figure 1]a and the docking structure 5NLX/CFF was used as the positive control during this research to understand this targeting mechanism. Based on the crystal structure of A2AR (PDB entry: 5MZP), caffeine was previously reported to interact with residues ILE 66, VAL 84, PHE 168, GLU 169, MET 177, LEU 249, ASN 253, MET 270, and ILE 274.[8],[21] The residues of the structure of A2AR (PDB entry: 5MZP) were renumbered to be able to compare with the structure of A2AR (PDB entry: 5NLX) used in our model. The corresponding residues are detected as ILE 75, VAL 93, PHE 177, GLU 178, MET 186, LEU 354, ASN 358, MET 375, and ILE 379. As the result of docking of ligands to receptor, it was determined that caffeine interacts with residues ILE 379, ASN 358, LEU 354, MET 186, and VAL 93, while zeatin interacts with residues MET 186, PHE 177, VAL 93, ALA 90, ILE 75, ALA 72, ILE 379, LEU 354, and ASN 358 based on our model. This finding showed that zeatin has the ability to form more interaction than caffeine in the same binding pocket. One hundred eighty-one contacts in total were detected for 5NLX/ZEA complex, while 156 contacts were detected for 5NLX/CFF complex. Seven weak polar contacts, 3 polar contacts, 2 hydrogen bonds, 3 weak hydrogen bonds, and 21 hydrophobic contacts were found for 5NLX/ZEA. The hydrogen bonds were formed with the aminoacid residues ASN 358, ALA 90, and ALA 72; polar contacts were formed with ALA 90, ILE 379, ILE 75, and VAL 93; and hydrophobic contacts were formed with MET 186, PHE 177, and LEU 354. For 5NLX/CFF complex, 5 weak polar contacts, 1 polar contact, 1 hydrogen bond, and 2 weak hydrogen bonds were found. The hydrogen bonds were formed with MET 186, ASN 358, and LEU 354; polar contacts were formed with ILE 379 and VAL 93. As A2AR is a transmembrane protein, both 5NLX/CFF and 5NLX/ZEA complexes were aligned into the membrane [Figure 1]b for the MD simulations.
Figure 1: Detection of protein–ligand interaction and membrane alignment before the MD simulations. (a) Interaction of adenosine A2A receptor with ligands (zeatin and caffein) after docking was detected using Arpeggio[23] and PyMOL[19] was used to visualize the interactions. Seven weak polar contacts (orange), 3 polar contacts (red), 2 hydrogen bonds (cyan), 3 weak hydrogen bonds (green), and 21 hydrophobic contact (magnetta) were found for 5NLX/zeatin. For 5NLX/Caffeine complex, 5 weak polar contacts (orange), 1 polar contact (red), 1 hydrogen bond (cyan), and 2 weak hydrogen bonds (green) were found. The other interactions that not mentioned here include aromatic contacts, carbonyl interactions, or other type of interactions. (b) Both 5NLX/zeatin and 5NLX/caffeine complexes were placed into membrane via membrane protein tutorial[24] Zeatin: ZEA, Caffeine: CFF

Click here to view


The hydrophobic pocket residues of A2AR were simulated for 50 ns to compare the adjacency of zeatin and caffeine. RMSD of protein backbone atoms of 5NLX/ZEA and 5NLX/CFF complexes for three production runs reached a steady plateau after 10 ns, indicating stable simulations for analysis [Figure 2]a. To explore whether if caffeine and zeatin were held in close proximity to the hydrophobic pockets, the fluctuations of the atoms of caffeine and zeatin were analyzed and the average of three production runs was taken for both 5NLX/ZEA and 5NLX/CFF complexes. Zeatin showed lower flexibility than caffeine with A2AR based on RMSF as an expected result [Figure 2]c because zeatin has the ability to form more interaction than caffeine, which reduces the flexibility of zeatin
Figure 2: Identification of 5NLX/zeatin and 5NLX/caffeine complexes during molecular dynamics simulations. (a) Root mean square displacement of backbone atoms of 5NLX/zeatin and 5NLX/caffeine complexes was plotted for three production runs. Root mean square displacement of backbone atoms for both complexes and for each run was calculated in 3 Å to detect the similarity between structures and production runs. The required stability was reached after 10 ns. (b) The atoms of ligands (zeatin and caffeine) were described in the image. (c) Root mean square fluctuation of the atoms of zeatin for 5NLX/zeatin (red) complex versus root mean square fluctuation of the atoms of caffeine for 5NLX/caffeine (blue) complex was analyzed to compare two structures in 50-ns molecular dynamics simulation. It is determined that zeatin (blue) showed lower flexibility than caffeine (red). (d) Radius of gyration which means the distrubition of molecules of all compound was measured for both 5NLX/zeatin (red) and 5NLX/caffeine (blue) complex and it showed that structure shape stay stable during 50-ns Molecular dynamics simulation. Å: Angstrom, C: Carbon, O: Oxygen, N: Nitrogen

Click here to view


The interactions between ligands (zeatin and caffeine) and A2AR are given [Table 1] during 50-ns MD simulation. While hydrogen bonds between A2AR and zeatin were mostly formed with the residues MET 375 and HIS 383, hydrogen bonds between A2AR and caffeine were mostly formed with residues ASN 358. Furthermore, the number of hydrogen bonds on frames is also given [Figure 3]. Hydrophobic interactions between A2AR and zeatin were mostly formed with the residues ILE 379, LEU 354, ALA 382, PHE 177, TYR 376, HIS 383, and ILE 75, while hydrophobic interctions between A2AR and caffeine were mostly formed with the residues PHE 177, GLU 178, ILE 379, and HIS 369. Salt bridges were only observed between A2AR and caffeine with the residues GLU 178. These residues were detected based on higher occupancy (>50%) between 999 frames during 50-ns MD simulations. It was determined that most of the residues interact with both zeatin and caffeine, but the occupancy between the frames is higher for zeatin and the interactions with the occupancy under 50% for both zeatin and caffeine were not analyzed in this study. For instance, interaction between zeatin and ILE 379 was observed with the percentage of 86.4% between frames for 5NLX/ZEA complex, while it was observed with the percentage of 62.8% for 5NLX/CFF complex. Salt bridges were not detected for 5NLX/ZEA complex during the analysis. These findings indicate that zeatin and caffeine display common residue interactions the same binding pocket.
Figure 3: The plots of hydrogen bonds for 5NLX/zeatin and 5NLX/caffeine complex. Hydrogen bond anlaysis was performed in Visual Molecular Dynamics[25] for three production runs based on frames in 3 Å. 5NLX/Zeatin complex contains more hydrogen bonds than 5NLX/Caffeine complex for each production run in 50-ns molecular dynamics simulation

Click here to view
Table 1: The summary of interactions between adenosine A2A receptor with ligands (zeatin and caffeine)

Click here to view


After the docking of both ligands (zeatin and caffeine) to A2AR (PDB entry: 5NLX), interacting residues were observed via Arpeggio[23] as in the table. For MD simulations, interacting residues were detecting between the frames by using VMD[25] and the residues that have higher occupancy (>50%) between 999 frames were chosen during the analysis of hydrogen bonds, hydrophobic bonds, and salt bridges.

After the detection of critical residues for both 5NLX/ZEA and 5NLX/CFF complexes, distance for 50 ns and RMSF for 20–50 ns were analyzed based on the critical residues. In distance analysis, zeatin stayed stable with the residues TYR 376, LEU 354, ILE 379, MET 375 [Figure 4], PHE 177 [Figure 5], ALA 382 [Figure 6], ILE 75, and HIS 383 during 50 ns, while caffeine is fluctuated after 20 ns. Caffeine and zeatin get closer during interaction with the residue GLU 178 [Figure 7] between 20 and 50 ns. For the residue ASN 358, zeatin and caffeine get closer between 20 and 30 ns, and then, caffeine moves away from the residue ASN 358. In addition, caffeine and zeatin get closer to HIS 369 between 30 and 50 ns. Therefore, it was expected that RMSF of this residues would be lower for 5NLX/ZEA complex than 5NLX/CFF complex because zeatin stays more stable during the interaction with these residues in 50-ns MD simulation. RMSF values based on each reasidue was calculated between 20 and 50 ns, but RMSF of these residues was found as similar for both 5NLX/ZEA and 5NLX/CFF complexes in our model
Figure 4: The comparison of interactions with the residue MET 375 for 5NLX/zeatin and 5NLX/caffeine complex via analysis of distance and root mean square fluctuation. A. Distance of zeatin (red) and caffeine (blue) to the residues MET 375 of adenosine A2A receptor receptor were measured via Visual Molecular Dynamics.[25] While zeatin stays stable in ~5 Å distance during 50-ns molecular dynamics simulation, caffeine moves away after 20 ns, from ~5 Å to ~15–20 Å. (b) Based on root mean square fluctuation of backbone atoms of the residue MET 375 for both 5NLX/zeatin (red) and 5NLX/caffeine (blue) complexes, the flexibility of the residue MET 375 was observed as similar for both 5NLX/caffeine and 5NLX/zeatin complexes. Distance analysis provides better understanding for the comparison of the effect of caffeine and zeatin on Adenosine A2A receptor, shows that zeatin stays longer with the residues caffeine interacts normally. (c) The image that was obtained from 422. Frame (21, 1. ns) of the first production run for 5NLX/zeatin and 327. Frame (16, 35. ns) for 5NLX/caffeine specifically shows the interaction with the residue MET 375 and zeatin. The interaction was highlighted as polar interaction (orange) with the distance 3.4 Å and hydrogen bond (cyan) with the distance 2.7 Å for 5NLX/Zeatin, weak hydrogen bonds (green) with distance 3.6 and 3.4 Å for 5NLX/Caffeine via PyMol.[17] Distance was also calculated to observe the interaction in all 50-ns molecular dynamics simulation as atom–atom. It was determined from selected images that the interaction performs between the O atoms of residue MET 375 and the C17 and O16 atoms of zeatin on the one hand, while the interaction performs between the CE and SD atom of residue MET 375 and the C8 and N9 atoms of caffeine on the other hand. The atom–atom distance graph showed that interaction stays stable in 5 Å distance for 5NLX/Zeatin complex, but distance was increasing after 20 ns for 5NLX/Caffeine complex. This result indicates that when zea interacts with the residue MET 375, CFF moves away from the interaction point. Zeatin: ZEA, Caffeine: CFF, Å: Angstrom, C: Carbon, CA (Cα): alpha carbon, CB (Cβ): Beta carbon, CG (Cγ): Gamma carbon, CE (Cε): Epsilon carbon O: Oxygen, SD (Sδ): Delta sülfür, N: Nitrogen

Click here to view
Figure 5: The comparison of interactions with the residue PHE 177 for 5NLX/zeatin and 5NLX/caffeine complex via analysis of distance and root mean square fluctuation. A. Distance of zeatin (red) and caffeine (blue) to the residues PHE 177 of adenosine A2A receptor was measured via visual molecular dynamics.[25] While zeatin stays stable in ~5–10 Å distance during 50-ns molecular dynamics simulation, caffeine moves away after 20 ns from 5 Å to ~10-15 Å. (b) Based on root mean square fluctuation of backbone atoms of the residue PHE 177 for both 5NLX/zeatin (red) and 5NLX/caffeine (blue) complexes, the flexibility of the residue PHE 177 was observed as almost similar for both 5NLX/caffeine and 5NLX/zeatin complexes. Distance analysis provides better understanding for the comparison of the effect of caffeine and zeatin on adenosine A2A receptor, shows that zeatin stays longer with the residues caffeine interacts normally. (c) The image that was obtained from 554. Frame (27.7. ns) of first production run for 5NLX/Zeatin and 544. Frame (27, 2. ns) for 5NLX/Caffeine specifically shows the interaction with the residue PHE 177 and zeatin. The interaction was highlighted as hydrophobic interaction (purple) with the distance 4.3 and 4.5 Å and weak hydrogen bond (green) with the distance 3.4 Å for 5NLX/zeatin, polar contacts (orange) with distance 3.4 and 3.5 Å for 5NLX/caffeine via PyMol.[17] Distance was also calculated to observe the interaction in all 50-ns molecular dynamics simulation as atom–atom. It was determined from selected images that the interaction performs between the CD1 and CE2 atoms of residue PHE 177 and the C12 and O16 atoms of zeatin on the one hand, while the interaction performs between the CD1 and CG atom of residue PHE 177 and the O11 atoms of caffeine on the other hand. The atom–atom distance graph showed that interaction stays stable in 5–10 Å for 5NLX/zeatin complex but distance was increasing after 20 ns for 5NLX/caffeine complex. This result indicates that when zeatin interacts with the residue, caffeine moves away from the interaction point. Zeatin: ZEA, Caffeine: CFF, Å: Angstrom, C: Carbon, CA (Cα): alpha carbon, CB (Cβ): Beta carbon, CG (Cγ): Gamma carbon, CD (Cδ): Delta carbon, CE (Cε): Epsilon carbon, CZ (Cζ): Zeta carbon, O: Oxygen, N: Nitrogen

Click here to view
Figure 6: The comparison of interactions with the residue ALA 382 for 5NLX/zeatin and 5NLX/caffeine complex via analysis of distance and root mean square fluctuation. (a) Distance of zatin (red) and caffeine (blue) to the residues ALA 382 of adenosine A2A receptor was measured via Visual Molecular Dynamics.[24] While zeatin stays stable in ~5-10 Å distance during 50-ns molecular dynamics simulation, caffeine moves away after 20 ns from 10 Å to ~20 Å. (b) Based on root mean square fluctuation of backbone atoms of the residue ALA 382, both 5NLX/zeatin (red) and 5NLX/caffeine (blue) complexes showed that flexibility of the residue ALA 382 is lower for 5NLX/zeatin as expected and this residue may play an important role in the stability of zeatin during the interaction with Adenosine A2A receptor. (c) The image that was obtained from 457. Frame (22, 85. ns) of the first production run for 5NLX/zeatin specifically shows the interaction with the residue ALA 382 and zeatin. The interaction was highlighted as hydrophobic interaction (purple) with the distance 3.8 Å for 5NLX/zeatin via PyMol.[17] Distance was also calculated to observe the interaction in all 50-ns molecular dynamics simulation as atom–atom. It was determined from selected images that the interaction performs between the CB atom of residue ALA 382 and the C15 atom of zeatin. No interaction was detected between caffeine and the residue ALA 382 during analysis with the occupancy >50%. The atom–atom distance graph showed that interaction stays stable in 5 Å until ~25 ns and after, distance reachs 10 Å. Zeatin: ZEA, Caffeine: CFF, Å: Angstrom, C: Carbon, CA (Cα): alpha carbon, CB (Cβ): Beta carbon, O: Oxygen, N: Nitrogen

Click here to view
Figure 7: The comparison of interactions with the residue GLU 178 for 5NLX/zeatin and 5NLX/caffeine complex via analysis of distance and root mean square fluctuation. (a) Distance of zeatin (red) and caffeine (blue) to the residues GLU 178 of adenosine A2A receptor measured via Visual Molecular Dynamics.[25] While zeatin stays stable in ~10-15 Å distance during 50-ns molecular dynamics simulation, caffeine stays stable until 20 ns then, it gets closer with GLU 178 between 20 and 50 ns, reaching ~15 Å. These data also indicate that caffeine and zeatin stay close position to each others after 20 ns. (b) Based on root mean square fluctuation of backbone atoms of the residue GLU 178, both 5NLX/zeatin (red) and 5NLX/caffeine (blue) complexes showed that flexibility of the residue GLU 178 was observed as similar for both 5NLX/caffeine and 5NLX/zeatin complexes. The possible interaction was formed with another residue that can reduce the the flexibility of GLU 178 for 5NLX/caffeine complex (c) The image that was obtained from 924. Frame (46, 2. ns) of first production run for 5NLX/Zeatin and 707. Frame (35, 35. ns) for 5NLX/caffeine specifically shows the interaction with the residue GLU 178 and zeatin. The interaction was highlighted as weak hydrogen bond (green) with the distance 3.6 and 3.5 Å for 5NLX/Zeatin and 3.8 Å for 5NLX/Caffeine via PyMol.[17] Distance was also calculated to observe the interaction in all 50 ns molecular dynamics simulation as atom–atom. It was determined from selected images that the interaction performs between the CB and N atoms of residue E178 and the O16 atom of zeatin on the one hand, while the interaction performs between the CB atom of residue GLU 178 and the O13 atom of caffeine on the other hand. The atom–atom distance graph showed that interaction stays stable in 10-15 Å for 5NLX/Zeatin complex, but the distance was increasing after 20 ns for 5NLX/caffeine complex and caffeine gets closer position with the residue and zeatin. This result indicates differently from other residues that both zeatin and caffeine stay stable with the residue GLU 178 after 20 ns which supports the previous root mean square fluctuation results. GLU 178 can play a critical role in stability of caffeine as a salt bridge. Zeatin: ZEA, Caffeine: CFF, Å: Angstrom, C: Carbon, CA (Cα): alpha carbon, CB (Cβ): Beta carbon, CG (Cγ): Gamma carbon, CD (Cδ): Delta carbon, O: Oxygen, OE (Oε): Epsilon oxygen, N: Nitrogen

Click here to view


The movement of ligands (zeatin and caffeine) was analyzed in each production run and it was detected that caffeine moves outward direction, while zeatin moves inward direction during 50-ns MD simulation [Figure 8]. The reason for this movement may be the formed new interactions with other residues in 5NLX/CFF complex during the simulation, leading the reduction of flexibility of the residues, but these all findings prove that zeatin has the ability to stay more stable during interaction with A2AR and interacts with either the same or different residues to stay longer in the binding pocket compared to caffeine. Zeatin may have a potential role as a ligand on A2AR and promising a new target for further studies
Figure 8: The movement of ligands (zeatin and caffeine) complex within 50-ns molecular dynamics simulation for three production runs. In 5NLX/zeatin complex, zeatin moves inward direction, while caffeine moved outward direction in 5NLX/caffeine complex during 50 ns and the most possible reason for the movement of caffeine is the formed new interactions between the different residues of protein and caffeine. Each color represents the structure in certain time. Structure in 0th ns: Green, Structure in 10th ns: Yellow, structure in 20th ns: pink, structure in 30th ns: cyan, structure in 40th ns: purple, structure in 50th ns: orange

Click here to view



  Discussion Top


A2ARs play a key role in regulating transmembrane signaling pathways in response to specific ligand and caffeine is known as one of the nonselective A2AR antagonists as well as it is the most consumed psychostimulant in the world.[33],[34] The main action of caffeine involves the blockade of A2AR on a various of physiological processes.[35] Despite of harmful effect of caffeine when it is overtaken, the main reason for consumption of caffeine is to in cognitive performance and mood.[36],[37] However, it enhances performance more in fatigued than well-rested subjects.[38],[39] It has recently been demonstrated that zeatin which is one of the plant hormones activates the mammalian A2AR, playing a role in the regulation of cells involved in both innate and adaptive immunities, as well as protects cognitive dysfunction such as improves memory impairment or mood disorders.[18],[40] The interaction mechanism of zeatin on A2AR has not been demonstrated clearly yet. The binding of ligand on A2AR causes the conformational changes on the receptor and leads the activation of signaling pathways.[8],[41] The model structure of A2AR in complex with zeatin and caffeine (as positive control) was constructed for this study and analyzed within 50-ns MD simulation with the repeated three production runs. Due to the size of the system, the simulation time was kept as 50 ns for each production run, but it can be extended for future studies to provide better estimation. The results indicate that both zeatin and caffeine have interaction in the same hydrophobic pocket. The residues TYR 376, LEU 354, ILE 379, MET 375, PHE 177, ALA 382, ILE 75, ASN 358, GLU178, HIS 383, and HIS 369 were determined as critical for the interaction between A2AR and both ligands. Residue-specific distance showed that zeatin stays more stable than caffeine during 50-ns MD simulation, supporting zeatin as a new target. Despite of the distance analysis, RMSF of each residue showed close results for both 5NLX/ZEA and 5NLX/CFF complexes. This finding provides a new question for the further studies, how the flexibility of the protein residues in 5NLX/CFF complex based on residue-specific RMSF analysis can be observed almost the same as in 5NLX/ZEA complex, although zeatin is determined as more stable based on distance analysis? In addition, zeatin was also determined as less flexible during RMSF analysis for ligands [Figure 2]. The most possible reason for it can be because of new performed interactions occurring on the specific residue with other residues of protein during 50-ns MD simulation. As the residues are allowed to interact for a period of time to observe dynamic evolution of the system, the flexibility of the specific residue can be observed as less for both the models. The combination of the study with distance analysis provided more accurate and sensitive results to detect the stability of zeatin on A2AR. Although zeatin was detected as stable in most of the choosen residues, the residue ALA 382 may play an important role in stabilization of zeatin in complex because the interaction with the residue ALA 382 only was performed in 5NLX/ZEA complex and the flexibility of the residue was lower for 5NLX/ZEA complex, indicating longer interaction during 50-ns MD simulation. During the study, local rigidity was considered to understand the stability of binding by looking RMSD, RMSF, and distance analysis, but these considerations were not considered enough to understand exactly the stability of binding and binding affinity. For further studies, binding energy estimation is required to detect the effect of zeatin on A2AR and the strength of the binding interaction. By this study, zeatin is identified as a potential ligand of A2AR based on computational model for the first time and the interacting resdiues were analyzed to provide a better understanding of the binding mechanism based on the positive control described as A2AR in complex with caffeine


  Conclusion Top


Zeatin, a plant hormone, has recently been detected as a potential target for A2AR, but protein–ligand interaction mechanism has not been clarified yet. Based on structural modeling, zeatin was identified as a ligand of A2AR for the first time and interacting resdiues were analyzed based on A2AR in complex with caffeine as a positive control to reveal the binding mechanism.

Patient informed consent

There is no need for patient informed consent.

Ethics committee approval

There is no need for ethics committee approval.

Financial support and sponsorships

No funding was received.

Conflicts of Interest

There is no conflicts of interest to declare.

Author contribution subject and rate:

Ebru Destan (%34): Collected data, performed analysis and wrote the paper.

Ahmet Can Timucin (%33): Conceived and designed the analysis, collected data and wrote the paper.

Pınar Öz (%33): Conceived and designed the analysis, collected data and wrote the paper.



 
  References Top

1.
Lazarus M, Shen HY, Cherasse Y, Qu WM, Huang ZL, Bass CE, et al. Arousal effect of caffeine depends on adenosine A2A receptors in the shell of the nucleus accumbens. J Neurosci 2011;31:10067-10075. [Doi:10.1523/JNEUROSCI.6730-10.2011].  Back to cited text no. 1
    
2.
Thiel M, Caldwell CC, Sitkovsky MV. The critical role of adenosine A2A receptors in downregulation of inflammation and immunity in the pathogenesis of infectious diseases. Microbes Infect 2003;5:515-26.  Back to cited text no. 2
    
3.
Schiffmann SN, Fisone G, Moresco R, Cunha RA, Ferré S. Adenosine A2A receptors and basal ganglia physiology. Prog Neurobiol 2007;83:277-92.  Back to cited text no. 3
    
4.
Doré AS, Robertson N, Errey JC, Ng I, Hollenstein K, Tehan B, et al. Structure of the adenosine A 2A receptor in complex with ZM241385 and the xanthines XAC and caffeine. Structure 2011; 19:1283-1293. [Doi: 10.1016/j.str. 2011.06.014].  Back to cited text no. 4
    
5.
IJzerman AP, van der Wenden EM, van Galen PJ, Jacobson KA. Molecular modeling of adenosine receptors. The ligand binding site on the rat adenosine A2A receptor. Eur J Pharmacol 1994;268:95-104.  Back to cited text no. 5
    
6.
Jaakola VP, Ijzerman AP. The crystallographic structure of the human adenosine A2A receptor in a high-affinity antagonist-bound state: İmplications for GPCR drug screening and design. Curr Opin Struct Biol 2010;20:401-14.  Back to cited text no. 6
    
7.
Fredholm BB, Chen JF, Masino SA, Vaugeois JM. Actions of adenosine at its receptors in the CNS: İnsights from knockouts and drugs. Annu Rev Pharmacol Toxicol 2005;45:385-412.  Back to cited text no. 7
    
8.
Carpenter B, Lebon G. Human Adenosine A2A Receptor: Molecular mechanism of ligand binding and activation. Front Pharmacol 2017;8:898.  Back to cited text no. 8
    
9.
Fredholm BB. Adenosine, adenosine receptors and the actions of caffeine. Pharmacol Toxicol 1995;76:93-101. [Doi: 10.1111/j. 1600-0773.1995.tb00111.x].  Back to cited text no. 9
    
10.
Dall'lgna OP, Porciúncula LO, Souza DO, Cunha RA, Lara DR. Neuroprotection by caffeine and adenosine A 2A receptor blockade of β-amyloid neurotoxicity. Br J Pharmacol 2003; 138:1207-1209. [Doi: 10.1038/sj.bjp.0705185].  Back to cited text no. 10
    
11.
Cornelis MC, El-Sohemy A, Campos H. Genetic polymorphism of the adenosine A2A receptor is associated with habitual caffeine consumption. Am J Clin Nutr 2007;86:240-4.  Back to cited text no. 11
    
12.
Dall'Igna OP, Fett P, Gomes MW, Souza DO, Cunha RA, Lara DR. Caffeine and adenosine A2a receptor antagonists prevent β-amyloid (25-35)-induced cognitive deficits in mice. Exp Neurol 2007;203:241-245. [Doi: 10.1016/j.expneurol.2006.08.008].  Back to cited text no. 12
    
13.
Lee YC, Yang YC, Huang CL, Kuo TY, Lin JH, Yang DM, et al. When cytokinin, a plant hormone, meets the adenosine A2A receptor: A novel neuroprotectant and lead for treating neurodegenerative disorders? PLoS One 2012;7:e38865. [Doi: 10.1371/journal.pone. 0038865].  Back to cited text no. 13
    
14.
Voller J, Maková B, Kadlecová A, Gonzalez G, Strnad M. Plant Hormone Cytokinins for Modulating Human Aging and Age-Related Diseases. 2017:311-335.  Back to cited text no. 14
    
15.
Choi SJ, Jeong CH, Choi SG, et al. Zeatin prevents amyloid β-induced neurotoxicity and scopolamine-induced cognitive deficits. J Med Food 2009;12:271-277. [Doi: 10.1089/jmf.2007.0678].  Back to cited text no. 15
    
16.
Öz P, Timuçin AC, Teomete Ş, Akpunar F, Tufanç Ç, Oğur D et al. The sex-dependent anti-depressant-like effects of zeatin in rat behavioral despair model as a candidate A2A receptor ligand. Neurosci Lett 2020;734:135108. [Doi: 10.1016/j.neulet.2020.135108].  Back to cited text no. 16
    
17.
Himer L, Csóka B, Selmeczy Z, Koscsó B, Pócza T, Pacher P, et al. Adenosine A2A receptor activation protects CD4+T lymphocytes against activation-induced cell death. FASEB J 2010;24:2631-2640. [Doi: 10.1096/fj.10-155192].  Back to cited text no. 17
    
18.
Lappas CM. The plant hormone zeatin riboside inhibits T lymphocyte activity via adenosine A2A receptor activation. Cell Mol Immunol 2015;12:107-12.  Back to cited text no. 18
    
19.
DeLano WL. The PyMOL Molecular Graphics System. Ver. 2.3. Schr{ö}dinger LLC; 2002.  Back to cited text no. 19
    
20.
Zoete V, Cuendet MA, Grosdidier A, Michielin O. SwissParam: A fast force field generation tool for small organic molecules. J Comput Chem 2011;32:2359-2368. [Doi: 10.1002/jcc.21816].  Back to cited text no. 20
    
21.
Cheng RK, Segala E, Robertson N, Deflorian F, Doré A, Errey JC, et al. Structures of Human A1 and A2A adenosine receptors with xanthines reveal determinants of selectivity. Structure 2017; 25:1275-1285.e4. [Doi: 10.1016/j.str.2017.06.012].  Back to cited text no. 21
    
22.
Goodsell DS, Morris GM, Olson AJ. Automated docking of flexible ligands: Applications of AutoDock. J Mol Recognit 1996;9:1-5. [Doi: 10.1002/(SICI)1099-1352(199601) 9:1<1:AID-JMR241>3.0.CO;2-6].  Back to cited text no. 22
    
23.
Jubb HC, Higueruelo AP, Ochoa-Montaño B, Pitt WR, Ascher DB, Blundell TL. Arpeggio: A web server for calculating and visualising ınteratomic ınteractions in protein structures. J Mol Biol 2017;429:365-371. [Doi: 10.1016/j.jmb.2016.12.004].  Back to cited text no. 23
    
24.
Aksimentiev A, Sotomayor M, Wells D, Huang Z. Membrane Proteins Tutorial. Theor Comput Biophys Group, Univ Illinois Urbana-Champaign; 2009.  Back to cited text no. 24
    
25.
Humphrey W, Dalke A, Schulten K. VMD: Visual molecular dynamics. J Mol Graph 1996;14:33-8, 27-8.  Back to cited text no. 25
    
26.
Phillips JC, Braun R, Wang W, Gumbart J, Tajkhorshid E, Villa E, et al. Scalable molecular dynamics with NAMD. J Comput Chem 2005;26:1781-1802. [Doi: 10.1002/jcc.20289].  Back to cited text no. 26
    
27.
MacKerell AD, Bashford D, Bellott M, Dunbrack RL, Evanseck JD, Field MJ, et al. All-atom empirical potential for molecular modeling and dynamics studies of proteins. J Phys Chem B 1998;102:3586-616.  Back to cited text no. 27
    
28.
Brooks BR, Brooks CL, Mackerell AD, Nilsson L, Petrella RJ, Roux B, et al. CHARMM: The biomolecular simulation program. J Comput Chem 2009;30:1545-1614. [Doi: 10.1002/jcc.21287].  Back to cited text no. 28
    
29.
Feig M, MacKerell AD, Brooks CL. Force field influence on the observation of π-helical protein structures in molecular dynamics simulations. J Phys Chem B 2003;107:2831-2836. [Doi: 10.1021/jp027293y].  Back to cited text no. 29
    
30.
MacKerell a D, Feig M, Brooks CL. Extending the treatment of backbone energetics in protein force fields. J Comp Chem 2004;25:1400-1415. [Doi: 10.1002/Jcc.20065].  Back to cited text no. 30
    
31.
Jorgensen WL, Chandrasekhar J, Madura JD, Impey RW, Klein ML. Comparison of simple potential functions for simulating liquid water. J Chem Phys 1983;79:926-935. [Doi: 10.1063/1.445869].  Back to cited text no. 31
    
32.
Miller JM. GraphPad PRISM. Analysis; 2003. Available from: http://www.graphpad.com.  Back to cited text no. 32
    
33.
Stevens RC, Cherezov V, Katritch V, Abagyan R, Kuhn P, Rosen H, et al. The GPCR Network: A large-scale collaboration to determine human GPCR structure and function. Nat Rev Drug Discov 2013;12:25-34.  Back to cited text no. 33
    
34.
Boia R, Ambrósio AF, Santiago AR. Therapeutic opportunities for caffeine and A2A receptor antagonists in retinal diseases. Ophthalmic Res 2016;55:212-8.  Back to cited text no. 34
    
35.
López-Cruz L, Salamone JD, Correa M. Caffeine and selective adenosine receptor antagonists as new therapeutic tools for the motivational symptoms of depression. Front Pharmacol 2018;9:526.  Back to cited text no. 35
    
36.
Smith AP. Caffeine, extraversion and working memory. J Psychopharmacol 2013;27:71-6.  Back to cited text no. 36
    
37.
Pasman WJ, Boessen R, Donner Y, Clabbers N, Boorsma A. Effect of caffeine on attention and alertness measured in a home-setting, using web-based cognition tests. JMIR Res Protoc 2017;6:e169.  Back to cited text no. 37
    
38.
Lorıst MM, Snel J, Kok A, Mulder G. Influence of caffeine on selective attention in well-rested and fatigued subjects. Psychophysiology 1994;31:525-534. [Doi: 10.1111/j. 1469-8986.1994.tb02345.x].  Back to cited text no. 38
    
39.
Childs E, de Wit H. Enhanced mood and psychomotor performance by a caffeine-containing energy capsule in fatigued individuals. Exp Clin Psychopharmacol 2008;16:13-21.  Back to cited text no. 39
    
40.
Kim MJ, Choi SJ, Lim ST, Kim HK, Kim YJ, Yoon HG et al. Zeatin supplement improves scopolamine-induced memory impairment in mice. Biosci Biotechnol Biochem 2008;72:577-581. [Doi: 10.1271/bbb.70480].  Back to cited text no. 40
    
41.
Dalton JA, Lans I, Giraldo J. Quantifying conformational changes in GPCRs: Glimpse of a common functional mechanism. BMC Bioinformatics 2015;16:124.  Back to cited text no. 41
    


    Figures

  [Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8]
 
 
    Tables

  [Table 1]



 

Top
 
 
  Search
 
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

 
  In this article
Abstract
Introduction
Materials and Me...
Results
Discussion
Conclusion
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed1908    
    Printed116    
    Emailed0    
    PDF Downloaded116    
    Comments [Add]    

Recommend this journal


[TAG2]
[TAG3]
[TAG4]