Synthesis, Biological Evaluation and Molecular Modeling Studies of Naphthoquinone Sulfonamides and Sulfonate Ester Derivatives as P2X7 Inhibitors

ATP acts in the extracellular environment as an important signal, activating a family of receptors called purinergic receptors. In recent years, interest in the potential therapeutics of purinergic components, including agonists and antagonists of receptors, has increased. Currently, many observations have indicated that ATP acts as an important mediator of inflammatory responses and, when found in high concentrations in the extracellular space, is related to the activation of the P2X7 purinergic receptor. In this sense, the search for new inhibitors for this receptor has attracted a great deal of attention in recent years. Sulfonamide derivatives have been reported to be potent inhibitors of P2X receptors. In this study, ten naphthoquinone sulfonamide derivatives and five naphthoquinone sulfonate ester derivatives were tested for their inhibitory activity on the P2X7 receptor expressed in peritoneal macrophages. Some compounds showed promising results, displaying IC50 values lower than that of A740003. Molecular docking and dynamic studies also indicated that the active compounds bind to an allosteric site on P2X7R. The binding free energy indicates that sulfonamides have an affinity for the P2X7 receptor similar to A740003. Therefore, the compounds studied herein present potential P2X7R inhibition.


Introduction
Purinergic receptors are transmembrane receptors widely expressed in mammalian tissues. These receptors are involved in several physiological and pathological processes, such as cell proliferation, migration and differentiation, cytokine production, pain, apoptosis, and tumorigenesis [1]. Initially, they were classified into two subgroups depending on their responsiveness to extracellular adenosine (P1 receptors) or purine/pyrimidine nucleotides (P2 receptors) [2,3]. P1 receptors are G protein protein-coupled receptors (GPCRs) that differ in the type of associated G protein. Four P1 receptor subtypes have been identified in vertebrate tissues (A 1 , A 2A , A 2B and A 3 ) [4]. The P2 receptor family is further subdivided into P2X and P2Y based on function and structural composition [5,6]. While P2Y receptors are also GPCRs, activated by ATP and purine/pyrimidine nucleotides,

Chemistry
The compounds were synthetized as reported in a previous study [28]. The chemical structures of the assessed compounds are shown in Figure 1.

Cytotoxicity Assay
Initially, we evaluated the toxicity of 15 naphthoquinone derivatives (PS01−15) in mouse peritoneal macrophages continuously exposed for 24 h. All compounds showed no significant cytotoxicity ( Figure 2) when compared with untreated cells (negative control). As expected, the detergent Triton X−100 impaired the resazurin reduction in treated cells. This absence of a PS series effect on cell metabolism motivated further investigations into their effect on P2X7 antagonism [29][30][31].

Dye Uptake Assay
Stimulation of peritoneal macrophages with ATP at millimolar concentrations induces the opening of pores permeable to fluorescent dyes, such as propidium iodide [32,33] Therefore, we preincubated these cells with a sulfonamide series for 5 min before ATP addition. Treatment with ATP induced submaximal PI uptake compared with the positive Molecules 2023, 28, 590 3 of 15 control, and A740003 inhibited the ATP effect. The substances PS04, PS05, PS06, PS07, PS08, PS11, PS12, PS13, PS14 and PS15 did not show inhibitory effects (at a concentration of 10 µM). PS01, PS02, PS03, PS09, and PS10 reduced the ATP effect by more than 50% ( Figure 3).

Cytotoxicity Assay
Initially, we evaluated the toxicity of 15 naphthoquinone derivatives (PS01−15) in mouse peritoneal macrophages continuously exposed for 24 h. All compounds showed no significant cytotoxicity ( Figure 2) when compared with untreated cells (negative control). As expected, the detergent Triton X−100 impaired the resazurin reduction in treated cells. This absence of a PS series effect on cell metabolism motivated further investigations into their effect on P2X7 antagonism [29][30][31].

% Fluorescent Cells
Although the selected sulfonamides showed negligible alterations in cellular metabolism (see Figure 2), we tested their toxicity with increasing concentrations in mouse peritoneal cells by measuring lactate dehydrogenase (LDH) activity. Once again, the CC50 values for all compounds indicated no toxicity [35,36]. We determined IC50 values of 201 ± 11 μM, 298 ± 16 μM, 322 ± 20 μM, 401 ± 18 μM, and 351 ± 21 μM for PS01, PS02, PS03, PS09, and PS10, respectively ( Figure 5).  Another hallmark of P2X7R activation in the inflammatory response is IL−1β release after ATP stimulus. Thus, we treated human THP-1 cells and mouse peritoneal macrophages with LPS for 4 h and added ATP for 30 min to increase IL−1β release. Pretreatment with BBG, A740003 and all selected sulfonamides reversed this effect. Interestingly, PS01, PS02, and PS03 were able to inhibit P2X7R-mediated IL−1β release at low micromolar concentrations (Table 1). This inhibition was higher than BBG inhibition and worse than A740003. PS09 and PS10 exhibited higher inhibitory activity (Table 1). When compared with the ATP−induced pore formation assay, we observed an inversion of potency inhibition order. The distinct intracellular signaling pathways associated with Another hallmark of P2X7R activation in the inflammatory response is IL−1β release after ATP stimulus. Thus, we treated human THP-1 cells and mouse peritoneal macrophages with LPS for 4 h and added ATP for 30 min to increase IL−1β release. Pretreatment with BBG, A740003 and all selected sulfonamides reversed this effect. Interestingly, PS01, PS02, and PS03 were able to inhibit P2X7R-mediated IL−1β release at low micromolar concentrations (Table 1). This inhibition was higher than BBG inhibition and worse than A740003. PS09 and PS10 exhibited higher inhibitory activity (Table 1). When compared with the ATP−induced pore formation assay, we observed an inversion of potency inhibition order. The distinct intracellular signaling pathways associated with the P2X7R pore and P2X7R/inflammasome/IL−1β pathways might justify this discrepancy [33].
To evaluate the anti-inflammatory effects of the selected sulfonamides in vivo, we performed an ATP−induced edema formation assay. Treatment with all selected sulfonamides inhibited ATP−induced edema formation with ID50 values of 128 ± 16 ng/kg, 93 ± 8 ng/kg, 67 ± 5 ng/kg, 9.8 ± 0.7 ng/kg, and 13.6 ± 16 ng/kg for PS01, PS02, PS03, PS09, and PS10, respectively ( Figure 6). Interestingly, the anti-inflammatory effect of selected sulfonamides revealed PS09 and PS10 as the most promising P2X7R antagonists, although all derivatives exhibited potent inhibitory effects. Figure 6. Anti−edematogenic effects of sulfonamides on ATP-induced paw edema in mice. Groups with five Swiss Webster mice treated with ATP (intra-plantar) were preincubated for 30 min with increasing doses of sulfonamides. All sulfonamide data were compared with crescent BBG doses (data not shown). Paw edema was measured at 30 min after ATP application. These results represent three distinct days with five animals for each group.

Molecular Docking and Dynamics
The molecular docking results indicated a similar binding conformation for the sulfonamide compounds into the P2X7 allosteric pocket. The 1,4-naphthoquinone moiety of the sulfonamide compounds superposed well with the quinolinyl moiety of A740003 from the crystal structure (PDB: 5U1U). Thus, both moieties showed hydrophobic interactions with the residues Phe108, Val312, Met105, and Phe88. In addition, the sulfonamide compounds also presented interactions with residues located deep within the hydrophobic cavity, such as Tyr295, Phe95, Phe293, and Thr94. These interactions were also reported in the co-crystallization work of Karasawa and Kawate [37]. Furthermore, the 1,4-naphthoquinone core was already suggested as a pharmacophore Figure 6. Anti−edematogenic effects of sulfonamides on ATP-induced paw edema in mice. Groups with five Swiss Webster mice treated with ATP (intra-plantar) were preincubated for 30 min with increasing doses of sulfonamides. All sulfonamide data were compared with crescent BBG doses (data not shown). Paw edema was measured at 30 min after ATP application. These results represent three distinct days with five animals for each group.

Molecular Docking and Dynamics
The molecular docking results indicated a similar binding conformation for the sulfonamide compounds into the P2X7 allosteric pocket. The 1,4-naphthoquinone moiety of the sulfonamide compounds superposed well with the quinolinyl moiety of A740003 from the crystal structure (PDB: 5U1U). Thus, both moieties showed hydrophobic interactions with the residues Phe108, Val312, Met105, and Phe88. In addition, the sulfonamide compounds also presented interactions with residues located deep within the hydrophobic cavity, such as Tyr295, Phe95, Phe293, and Thr94. These interactions were also reported in the cocrystallization work of Karasawa and Kawate [37]. Furthermore, the 1,4-naphthoquinone core was already suggested as a pharmacophore group for P2X7 allosteric inhibitors by our previous work [38]. Figure 7 presents the superposition of the sulfonamide compounds docked into the allosteric site, as well as the A740003 ligand. group for P2X7 allosteric inhibitors by our previous work [38]. Figure 7 presents the superposition of the sulfonamide compounds docked into the allosteric site, as well as the A740003 ligand. To assess the sulfonamide compound stability and binding free energy into the P2X7 allosteric site, we performed molecular dynamics simulations of 50 ns for P2X7 in complex with each of the top five sulfonamides and, for comparison, the A740003 ligand. All 6 ligands (A740003, PS01, PS02, PS03, PS09, and PS10) complexed with the P2X7 receptor presented a stable trajectory conformation (Supplementary Materials, Figure S1). The ligand RMSD variation in relation to the P2X7 receptor indicated that the ligands were accommodated in the pocket at the beginning of the simulation and remained stable until the end (Figure 8). Although compound PS03 accommodated approximately 40-50 ns, it also remained stable in the allosteric site until the end of the simulation.  To assess the sulfonamide compound stability and binding free energy into the P2X7 allosteric site, we performed molecular dynamics simulations of 50 ns for P2X7 in complex with each of the top five sulfonamides and, for comparison, the A740003 ligand. All 6 ligands (A740003, PS01, PS02, PS03, PS09, and PS10) complexed with the P2X7 receptor presented a stable trajectory conformation (Supplementary Materials, Figure S1). The ligand RMSD variation in relation to the P2X7 receptor indicated that the ligands were accommodated in the pocket at the beginning of the simulation and remained stable until the end (Figure 8). Although compound PS03 accommodated approximately 40-50 ns, it also remained stable in the allosteric site until the end of the simulation. group for P2X7 allosteric inhibitors by our previous work [38]. Figure 7 presents the superposition of the sulfonamide compounds docked into the allosteric site, as well as the A740003 ligand. To assess the sulfonamide compound stability and binding free energy into the P2X7 allosteric site, we performed molecular dynamics simulations of 50 ns for P2X7 in complex with each of the top five sulfonamides and, for comparison, the A740003 ligand. All 6 ligands (A740003, PS01, PS02, PS03, PS09, and PS10) complexed with the P2X7 receptor presented a stable trajectory conformation (Supplementary Materials, Figure S1). The ligand RMSD variation in relation to the P2X7 receptor indicated that the ligands were accommodated in the pocket at the beginning of the simulation and remained stable until the end (Figure 8). Although compound PS03 accommodated approximately 40-50 ns, it also remained stable in the allosteric site until the end of the simulation.  After completing the 50 ns of simulation, all sulfonamide compounds slightly moved toward the inner of the P2X7 ion channel pore, becoming more buried into the allosteirc site. Therefore, the sufonamide compounds had a high affinity for this cavity. By analyzing the last frame from the molecular dynamics for each complex, we observed that the 1,4-naphthoquinone core of the sulfonamide compounds interacted mainly with the hydrophobic lateral chain of Val312, Ile310, Met105, and Phe88. In addition, PS01, PS02, and PS03 established a hydrogen bond between their secondary amine and the oxygen from Ala91 or Thr94. Figure 9 presents the sulfonamide compound interactions extracted from the last frame of the molecular dynamics. It is important to note that the addition of a small hydrophobic group at the para position on the benzenesulfonamide could contribute to a favorable binding affinity, since the compounds could move deeper into the cavity and perform hydrophobic interactions with residues located in deeper positions, such as Phe95, from the three different subunits, and Tyr295. On the other hand, compound PS09 presented an extra ring at this position that prevented it from performing such an interaction. Additionally, the flourination at this site, e.g., in PS02, was an interesting approach, not only for the hydrophobic contribution, but also for preventing the metabolic reaction of fase I at this site. Compound PS10 presented a NO 2 moiety at this site, and the electron-withdrawing effect caused by the nitro group in the aromatic ring changed the polarity of the compound, reducing the negative quadrupole of the aromatic ring. This fact favored parallel π-stacking conformations with the lateral chain from Tyr295. After completing the 50 ns of simulation, all sulfonamide compounds slightly moved toward the inner of the P2X7 ion channel pore, becoming more buried into the allosteirc site. Therefore, the sufonamide compounds had a high affinity for this cavity. By analyzing the last frame from the molecular dynamics for each complex, we observed that the 1,4-naphthoquinone core of the sulfonamide compounds interacted mainly with the hydrophobic lateral chain of Val312, Ile310, Met105, and Phe88. In addition, PS01, PS02, and PS03 established a hydrogen bond between their secondary amine and the oxygen from Ala91 or Thr94. Figure 9 presents the sulfonamide compound interactions extracted from the last frame of the molecular dynamics. It is important to note that the addition of a small hydrophobic group at the para position on the benzenesulfonamide could contribute to a favorable binding affinity, since the compounds could move deeper into the cavity and perform hydrophobic interactions with residues located in deeper positions, such as Phe95, from the three different subunit,s and Tyr295. On the other hand, compound PS09 presented an extra ring at this position that prevented it from performing such an interaction. Additionally, the flourination at this site, e.g., in PS02, was an interesting approach, not only for the hydrophobic contribution, but also for preventing the metabolic reaction of fase I at this site. Compound PS10 presented a NO2 moiety at this site, and the electron-withdrawing effect caused by the nitro group in the aromatic ring changed the polarity of the compound, reducing the negative quadrupole of the aromatic ring. This fact favored parallel π-stacking conformations with the lateral chain from Tyr295. The binding free energy indicated that sulfonamides had an affinity for the P2X7 receptor comparable to that of the known inhibitor A740003 (Figure 10). Moreover, PS10 presented a slightly more favorable binding free energy than A740003, which might be due to PS10 being more buried inside the P2X7 pore, which allowed this ligand to carry out hydrophobic interactions with residues from the three subunits located deep in the allosteric site. In addition, a strong staggered π-stacking interaction formed among The binding free energy indicated that sulfonamides had an affinity for the P2X7 receptor comparable to that of the known inhibitor A740003 (Figure 10). Moreover, PS10 presented a slightly more favorable binding free energy than A740003, which might be due to PS10 being more buried inside the P2X7 pore, which allowed this ligand to carry out hydrophobic interactions with residues from the three subunits located deep in the allosteric site. In addition, a strong staggered π-stacking interaction formed among Tyr295, the PS10 phenyl moiety and Phe95, stabilizing the PS10 conformation and contributing to a low binding free energy. Tyr295, the PS10 phenyl moiety and Phe95, stabilizing the PS10 conformation and contributing to a low binding free energy.

Chemistry
The naphthoquinone sulfonamide derivatives (PS01 to PS10) and sulfonate ester derivatives (PS11 to PS15) tested in this work were synthetized as previously reported [28].

Peritoneal Macrophage Isolation and Culture
The animals used in these experiments were furnished by the animal breeding unit at Pavilion Hélio and Peggy Pereira at FIOCRUZ/IOC. The Institutional Animal Ethics Committee approved all experimental protocols under number L039-2016. Peritoneal macrophages were isolated from euthanized mice (weighing between 20 and 30 g) by injection of 10 mL phosphate-buffered saline (PBS) into the peritoneum of each animal. Then, the solution was completely removed from the peritoneal cavity using a syringe and transferred to 50 mL sterile tubes. Tubes were centrifuged at 1500 rpm for 5 min. The supernatant was discarded, and the pellet was resuspended in RPMI medium supplemented with 10% SFB and 1% antibiotic. The cells were counted and seeded in plates. After 40 min, the supernatant from each well was discarded to remove cells other than macrophages, and then RPMI medium was added. The macrophages were maintained in RPMI medium supplemented with 10% SFB and 1% antibiotic in a humid atmosphere at 37 °C containing 5% CO2 during the experiments.

Resazurin Reduction Assay
For cell viability estimation, the resazurin reduction assay, commercially known as Alamar blue [39], was used. Briefly, the cells were seeded in 96-well plates at a concentration of 3.0 × 10 5 cells/well and kept in an oven at 37 °C under a 5% CO2 atmosphere. Then, the cultures were treated with 10 μM of each naphthoquinone derivative for 24 h. To reveal the assay, the treated cells were incubated for 4 h with resazurin, and the fluorescence of the material was determined using a SpectraMax M5 spectrophotometer (Molecular Devices, San Jose, CA, USA) at wavelengths of 570/595 nm

Chemistry
The naphthoquinone sulfonamide derivatives (PS01 to PS10) and sulfonate ester derivatives (PS11 to PS15) tested in this work were synthetized as previously reported [28].

Peritoneal Macrophage Isolation and Culture
The animals used in these experiments were furnished by the animal breeding unit at Pavilion Hélio and Peggy Pereira at FIOCRUZ/IOC. The Institutional Animal Ethics Committee approved all experimental protocols under number L039-2016. Peritoneal macrophages were isolated from euthanized mice (weighing between 20 and 30 g) by injection of 10 mL phosphate-buffered saline (PBS) into the peritoneum of each animal. Then, the solution was completely removed from the peritoneal cavity using a syringe and transferred to 50 mL sterile tubes. Tubes were centrifuged at 1500 rpm for 5 min. The supernatant was discarded, and the pellet was resuspended in RPMI medium supplemented with 10% SFB and 1% antibiotic. The cells were counted and seeded in plates. After 40 min, the supernatant from each well was discarded to remove cells other than macrophages, and then RPMI medium was added. The macrophages were maintained in RPMI medium supplemented with 10% SFB and 1% antibiotic in a humid atmosphere at 37 • C containing 5% CO 2 during the experiments.

Resazurin Reduction Assay
For cell viability estimation, the resazurin reduction assay, commercially known as Alamar blue [39], was used. Briefly, the cells were seeded in 96-well plates at a concentration of 3.0 × 10 5 cells/well and kept in an oven at 37 • C under a 5% CO 2 atmosphere. Then, the cultures were treated with 10 µM of each naphthoquinone derivative for 24 h. To reveal the assay, the treated cells were incubated for 4 h with resazurin, and the fluorescence of the material was determined using a SpectraMax M5 spectrophotometer (Molecular Devices, San Jose, CA, USA) at wavelengths of 570/595 nm (excitation/emission). For the positive control of the reaction, 0.5% Triton-X 100 surfactant (Sigma-Aldrich, St. Louis, MO, USA) was used.

LDH Release Assay
Peritoneal macrophages (3.0 × 10 5 cells/well) were grown in 96-well plates and kept in an oven at 37 • C under a 5% CO 2 atmosphere. The culture was treated with 10 µM of the substances for 24 h. The assay was finished by the addition of the LDH kit's reagent (Kit CytoTox 96®Non-Radioactive Cytotoxicity Assay, Promega, Madison, WI, USA), prepared according to the manufacturer's instructions. After 30 min of reaction, a stop solution was added, and the absorbance was measured at 490 nm.

Dye Uptake Assay
To evaluate P2X7R-associated pore formation, a propidium iodide (PI) uptake assay was used. Peritoneal macrophages (3.0 × 10 5 /well) were incubated with naphthoquinone derivatives and the inhibitor BBG (Brilliant Blue G, Sigma-Aldrich) at 10 µM and 750 nM, respectively, for 10 min. Subsequently, cells were exposed to 5 mM ATP for 15 min, and in the last 5 min, they were incubated with 1 µM PI dye. At the end of the incubation period, fluorescence was measured using a SpectraMax M5 spectrophotometer at 493/636 nm (excitation/emission). Compounds with inhibition values higher than 50% were selected to evaluate their inhibitory activity at varying concentrations. The inhibition curve was obtained using compounds at concentrations of 0.01 µM, 1 µM, 10 µM, 50 µM, and 100 µM. The controls and the procedure were the same.

IL-1β Enzyme-Linked Immunosorbent Assay
The monocytic THP-1 cell line used in this experiment was donated by the Laboratory of Applied Pharmacology, Institute of Drug Technology (Farmanguinhos), Rio de Janeiro, Brazil. Cells were plated at 2 × 10 5 cells/well in 96-well culture plates maintained in RPMI containing 10% fetal bovine serum (FBS), penicillin (100 U/mL), and streptomycin (100 mg/mL) in a humidified 5% CO 2 atmosphere at 37 • C. Then, the cells were primed with Escherichia coli lipopolysaccharide (LPS, serotype 0127: B8) four hours before ATP stimulation for 30 min. Mouse peritoneal macrophages were primed with 10 ng/mL interferon-γ (IFN-γ) for 24 h and posteriorly stimulated with Escherichia coli lipopolysaccharide (LPS, serotype 0127: B8). The P2X7R antagonist A740003 and naphthoquinone derivatives were preincubated for 30 min before ATP addition. IL-1β was determined according to a standard kit (Bioscience, San Diego, CA, USA and Abcam, Cambridge Biomedical Campus, Cambridge, UK).

ATP-Induced Paw Edema Assay
Paw edema in mice was induced by intra-plantar administration of ATP (10 mg/paw) for 30 min. Animals were pretreated intraperitoneally with the standard P2X7R antagonist A740003 (1 mg/kg) and the selected compounds PS01, PS02, PS03, PS09, and PS10 (1 mg/kg) for 60 min before ATP administration. Edema formation was determined with a mouse plethysmometer (UGO Basile, Gemonio, Italy) before ATP injection compared with saline injection in the contralateral paw edema.

Ligands and Receptor Modeling
Initially, the compounds were built in the program Avogadro [23,39], and then the molecular geometries were optimized using the MMFF94 force field [40]. Finally, the optimized conformers were submitted to refinement calculations of the optimized geometry through the semiempirical PM6 method performed by MOPAC2016 [41][42][43]. The threedimensional structure of the human P2X7R ion channel model was built as described in Faria et al., 2018 [38]. The primary sequence was obtained from the UniProtKB database (code: Q99572) [44]. As in this work we studied molecules with allosteric antagonist activity (PS01, PS02, PS03, PS09, and PS10), we used the crystallographic structure deposited under the PDB code of 5U1U as the template, since this structure is bound to the ligand A740003, which is a known allosteric inhibitor [37]. The ClustalW program (EMBL-EBI, Cambridgeshire, UK) was used for sequence alignment (Supplementary Materials, Figure S1) [45]. The ProsA program (Insoft Oy, Kempele, Finland) was used to assess the Z score and local model quality ( Figure S3 and Figure S4), and the PROCHECK program (EMBL-EBI) was used to generate the Ramachandran plot ( Figure S5) [46,47].

Molecular Docking
The ligand and P2X7 model were previously prepared using AutoDock tools 4.2.6 (CCSB, La Jolla, CA, USA) in which hydrogen atoms were added, as well as Gasteiger partial charges [48]. As the biological analysis indicated potential allosteric antagonist activity for the molecules, we placed the grid box around the allosteric site. The allosteric site comprised a groove located in the upper body domains of the P2X7 neighboring subunits. The main residues involved in the interaction with the allosteric antagonist were Phe95, Phe103, Met105, Phe293, and V312 [37]. Molecular docking was performed using the AutoDockVina program (CCSB, La Jolla, CA, USA) [49]. The exhaustiveness parameter was set to 10, and the energy range parameter was set to 30. This molecular docking approach was previously supported by redocking with the ligand A740003, described in Faria et al., 2018 [38].

Molecular Dynamics
Molecular dynamics was performed to evaluate the complex stability generated by the molecular docking process. The ligand force field parameters were previously generated by the Antechamber program [50]. The tLeap (Amber 18.0 Module) was used to generate the molecular system topology (ligand and protein). For the protein and ligands, the force field implemented was the FF14SB and GAFF (Generalized Amber Force Field), respectively [51]. The water model applied was TIP3P [52]. The water molecules were arranged around the solute at 1.5 Å and with a distance from the border of the simulation box of 15 Å. Additional ions (Na + and Cl − ) were added to neutralize the molecular system charges. Finally, the systems were minimized using the steepest descent method followed by the conjugate gradient method with 5000 iteration cycles. The particle-mesh-Ewald method was used for the treatment of long-range interactions, and a 9 Å limit was selected to compute the nonbonded interactions. The simulations were carried out by AMBER 18.0 with the PMEMD version (Particle Mesh Ewald Molecular Dynamics). Each complex system was thermalized with gradual heating of 30 K every 10 ps until reaching 300 K. The thermalization was performed in a canonical ensemble employing the Langevin thermostat. Next, the systems were equilibrated for 10 ps at 300 K under 1 atm pressure employing the Berendesen barostat [53]. Finally, molecular dynamics production was performed over 50 ns for each complex, with a time step for integration of 2 fs under canonical ensemble and periodic boundary conditions. The SHAKE algorithm was employed on atoms covalently bonded to a hydrogen atom. Root-mean-square deviations (RMSDs) were calculated to analyze the molecular dynamics trajectories. We performed the following two different RMSD measurements: one for the whole system (receptor and ligand) to evaluate the overall system stability, and the other for the ligand alone to evaluate the ligand trajectory and its stability in the binding pocket. However, in this case, we applied the "nofit" function to avoid superimposing the ligand structures onto each other prior to calculating RMSD, thus harnessing the movement of the inhibitors in relation to the receptor. Trajectory visual inspections were performed using the VMD program (Visual Molecular Dynamics version 1.9.2, Urbana, IL, USA).

Binding Free Energy Determination
To evaluate the stability of each complex and compare the sulfonamide affinity with the inhibitor A740003 at the P2X7 receptor, we performed binding free energy calculations. The binding free energy was calculated over the last 5 ns of simulation for each complex system. The generalized born method was applied by the MMPBSA.py script available in the AMBER 18 package [54].

Statistical Analysis
PRISM ® software (GraphPad, Inc., San Diego, CA, USA) was used for all statistical analyses. The measurements were performed in triplicate and, at least, on three independent days. All data are presented as the mean ± standard deviation of the means. Analysis of variance (ANOVA), followed by Tukey's test, was applied to all data. Values of p < 0.05 were considered statistically significant.

Conclusions
Ionotropic P2X7R has been implicated in the pathology of several inflammatory and noninflammatory diseases [55][56][57][58][59][60][61]. However, to date, no approved drugs targeting the P2X7 receptor have advanced to clinical usage, mainly due to a lack of selectivity. Our group has devoted great effort to the identification of novel chemical scaffolds to selectively target the P2X7 receptor. In recent years, we reported different groups of molecules with promising activity as P2X7 receptor inhibitors. Several synthetically modified naphthoquinones proved to be good molecular platforms for P2X7 receptor antagonists, as validated by in vitro, in vivo and in silico studies.
Herein, we reported the identification of a group of naphthoquinone derivatives with promising inhibitory activity against the P2X7 receptor. Among 15 compounds, we identified 5 naphthoquinone sulfonamide derivatives (PS01, PS02, PS03, PS09, PS10) with low toxicity against mammalian cells and better performance in the inhibition of ATP-induced dye uptake and P2X7 receptor-mediated IL-1 release than well-known P2X7 receptor antagonists. These compounds also presented promising anti-edematogenic effects of sulfonamides in an ATP-induced paw edema model. The molecular docking and dynamics results indicated that all the analyzed molecules were capable of interacting with the same residues involved with the known allosteric antagonists. The binding free energy indicated that sulfonamides have an affinity for the P2X7 receptor similar to the known inhibitor A740003. In addition, PS10 presented a more favorable binding free energy than A740003. All the results presented here support these compounds as being valuable synthetic platforms for obtaining novel and more selective P2X7R antagonists.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/molecules28020590/s1, Figure S1: RMSD variation of the ligands complexed with the P2X7 receptor during 50 ns of molecular dynamics simulation; Figure S2: Sequence alignment between the target P2X7 sequence and the template generated by the ClustalW program; Figure S3: Overall model quality of the P2X7 model generated by the ProSA; Figure S4: Local model quality of the P2X7 model generated by the ProSA; Figure S5: Ramachandran plot generated by the PROCHECK program.
Author Contributions: R.X.F. performed the biological assays. N.L.v.R. and C.R.R. performed the in-silico assays. P.A.F.P., D.T.G.G., D.R.d.R., F.d.C.d.S. and V.F.F. were involved in the planning and synthesis of the molecules. R.X.F. and N.L.v.R. prepared the figures, coordinated the synthesis assays, wrote, and revised the paper. P.A.F.P. and R.X.F. revised the biological assays and wrote and revised the paper. All authors have read and agreed to the published version of the manuscript.