Preparation and Evaluation of 6-Gingerol Derivatives as Novel Antioxidants and Antiplatelet Agents

Ginger (Zingiber officinale) is widely used as a spice and a traditional medicine. Many bioactivities have been reported for its extracts and the isolated compounds, including cardiovascular protective effects. Different pathways were suggested to contribute to these effects, like the inhibition of platelet aggregation. In this study, we synthesised fourteen 6-gingerol derivatives, including eight new compounds, and studied their antiplatelet, COX-1 inhibitor, and antioxidant activities. In silico docking of selected compounds to h-COX-1 enzyme revealed favourable interactions. The investigated 6-gingerol derivatives were also characterised by in silico and experimental physicochemical and blood–brain barrier-related parameters for lead and preclinical candidate selection. 6-Shogaol (2) was identified as the best overall antiplatelet lead, along with compounds 3 and 11 and the new compound 17, which require formulation to optimize their water solubility. Compound 5 was identified as the most potent antioxidant that is also promising for use in the central nervous system (CNS).


Introduction
Ginger, Zingiber officinale Rosc. (Zingiberaceae), is a well-known culinary plant and herbal remedy for many diseases. Ginger root extracts have been reported efficient against nausea and vomiting, diarrhoea, inflammatory conditions, metabolic syndrome, hepatotoxicity, and cardiovascular diseases [1][2][3]. The plant is rich in bioactive secondary metabolites including terpenes and phenolic compounds. The latter group consists mainly of gingerols, shogaols, paradols, and zingerone [4,5], among which 6-gingerol has been reported to be the most abundant in fresh ginger roots [6].
(Gilson, Middleton, WI, USA) with dual-wavelength detection, with an adequately chosen combination of acetonitrile-water, and a flow rate of 15 mL/min. The purity of the compounds obtained was determined by RP-HPLC analyses on a system of two Jasco PU 2080 pumps, a Jasco AS-2055 Plus intelligent sampler connected to a JASCO LC-Net II/ADC equipped with a Jasco MD-2010 Plus PDA detector (Jasco International Co. Ltd., Hachioji, Tokyo, Japan) utilizing a Kinetex C-18 (5 µm, 250 × 4.6 mm) column (Phenomenex Inc., Torrance, CA, USA) and applying a gradient of 30%-100% aqueous AcN in 30 min followed by 100% ACN for 10 min with a flow rate of 1 mL/min. Analysis of samples from the PAMPA and kinetic solubility assays was performed the same way, by using 3-point calibrations and integrating each compound at its UV absorption maximum. 1 H-and 13 C NMR spectra were recorded in CDCl 3 or CD 3 OD using 5 mm tubes at room temperature on a Bruker DRX-500 spectrometer at 500 ( 1 H) and 125 ( 13 C) MHz with the deuterated solvents' signal taken as reference. The heteronuclear single quantum coherence (HSQC), heteronuclear multiple bond correlation (HMBC), 1 H-1 H correlation spectroscopy (COSY), and nuclear Overhauser effect spectroscopy (NOESY) spectra were obtained using the standard Bruker pulse programs. 1  All bioactivity data processing, including the calculation of inhibition percentage, mean and corresponding standard error of the mean (SEM), and IC 50 values, was performed by GraphPad Prism 8.0 (La Jolla, CA, USA). IC 50 values were calculated from the sigmoidal dose-response curves obtained by the log(inhibitor) vs. response and variable slope (DPPH assay) or the log(inhibitor) vs. normalised response and variable slope nonlinear regression model. On the results obtained, no statistical evaluation was performed; instead, differences greater than two-fold were considered relevant. Plotting of the antiplatelet and COX-1 inhibitory IC 50 values and the linear regression of the data was performed by Microsoft Excel.

Antiplatelet Activity
Human platelet suspension (3 × 10 8 /mL in Tyrode's buffer) was prepared as previously described [42]. The protocol for this study was approved by the institutional review board of Kaohsiung Medical University Hospital (Kaohsiung City, Taiwan). Platelets pre-treated with DMSO (vehicle control) or test compounds were stimulated with arachidonic acid (AA), and platelet aggregation was measured using turbidimetric aggregometer (Chrono-Log Co., Havertown, PA, USA) at 37 • C under stirring conditions (1200 rpm).

Oxygen Radical Absorbance Capacity (ORAC)
AAPH ((2,20-Azobis(2-methyl-propionamidine) dihydrochloride) free radical and Trolox standard were purchased from Sigma-Aldrich Hungary. Fluorescein was purchased from Fluka Analytical, Tokyo, Japan. ORAC assay was carried out In a 96-well microplate based on the method from previous study [44]. Briefly, 20 µL of the samples (1 µM final concentration, dissolved in phosphate buffer, pH = 7.4, containing 1% MeOH) were mixed with 60 µL of AAPH (12 mM final concentration, dissolved in phosphate buffer, pH = 7.4) and 120 µL of fluorescein solution (70 nM final concentration, dissolved in phosphate buffer), then the fluorescence was measured (excitation at λ = 485 nm, and emission at λ = 520 nm) through 3 h with 1.5-min cycle intervals with a BMG Labtech FluoStar Optima plate-reader. All experiments were carried out in triplicate, and Trolox was used as standard. The antioxidant capacity is expressed as Trolox Equivalent (TE), as calculated using GraphPad Prism 8.0 (La Jolla, CA, USA).

Xanthine Oxidase Inhibitory Activity
The xanthine oxidase (XO) inhibitory activity of the compounds was determined using continuous spectrophotometric rate based on a modified protocol of Sigma. The samples were prepared in. In a 96-well plate, the final reaction mixture consisted of 10 µL of sample (dissolved in DMSO, 30 mM stock solution, 100 µM final), 100 µL of xanthine solution (0.15 mM, pH = 7.4), 140 µL of buffer (potassium phosphate, pH = 7.5) and 50 µL of XO (0.2 units/mL). When measuring the enzyme activity, control buffer was used in place of the sample. Allopurinol was applied as a control. The reaction was initiated by the automatic addition of 50 µL of XO solution. The absorbance of XO-induced uric acid production from xanthine was measured at 290 nm for 3 min in a 96-well plate on a BMG Labtech FluoStar Optima plate reader. The inhibitory percentage values were calculated by using Graph Pad Prism 8.0 (La Jolla, CA, USA).

Peroxynitrite Scavenging Activity
Peroxynitrite was synthesised by a continuous flow system using syringe pumps as published previously [45]. Briefly, an acidic solution of hydrogen peroxide (0.6 M H 2 O 2 , 0.7 M HCl) was pumped to a junction alongside of sodium nitrite solution (0.6 M) at a flow rate of 1.5 mL/min. After passing 10 cm of tubing, it was mixed with a sodium hydroxide solution (1.5 M), also pumped at 1.5 mL/min. The resulting peroxynitrite solution was a bright yellow colour. The tubing around the reaction was submerged in ice. The concentration of the solution was determined by spectrophotometry and was adjusted to 30 mM with 0.1 M NaOH solution.
In a 96-well microplate 245 µL of pyrogallol red (100 µM final concentration, dissolved in 0.1 M glycine buffer) was mixed with 50 µL of sample (0.5 mM final concentration, dissolved in DMSO) and 5 µL ONOOsolution (500 µM final concentration, freshly prepared). After mixing and keeping it at room temperature for 30 min the absorbance was measured at 550 nm using a FluoStar Optima plate reader (software version 2.20R2, BMG Labtech, Ortenberg, Germany).

COX-1 Inhibitory Activity
COX-1 inhibitory activity was tested based on the fluorometric method as described in BioVision's COX-1 inhibitor screening kit leafkit (K548-100, BioVision, CA, USA). Sample solutions were prepared by dissolving in DMSO and subsequently buffer, to get desired concentrations. In a 96-well white plate (655101, F-bottom, Grenier bio-one, Germany), 80 µL reaction mix (containing 76 µL assay buffer, 1 µL COX Probe, 2 µL COX cofactor, and 1 µL COX-1 enzyme) was added to 10 µL sample solution, DMSO and assay buffer to get test wells assigned for sample screen (S), negative control (N) and blank, respectively. An aliquot of 10 µL of arachidonic/NaOH solution was added to each well using a multichannel pipette to initiate the reaction at the same time, and the fluorescence of each well was measured kinetically at Ex/Em 550/610 nm, at 25OC for 10 min using a FluoStar Optima plate reader (BMG Labtech, Ortenberg, Germany). The COX inhibitory activity of SC560, a standard inhibitor, was also determined.
The change in fluorescence between two points, T1 and T2 were determined, and relative inhibition was calculated according to the following equation: where N is the absorbance of the negative control, and S is that of the sample. Dose-effect studies on the compounds were used to determine the concentration that inhibits 50% of the enzyme activity. The sigmoidal dose-response model was obtained by using the software GraphPad Prism 8.0 (La Jolla, CA, USA), and these were used to determine the IC 50 values of the compounds.

Physicochemical Character and Blood-brain Barrier Specific Permeability
Basic physicochemical parameters for drug design and candidate selection were calculated by Percepta Software Package (ACD/Labs, Toronto, Ontario, Canada) [46]. Tautomers and it is distributions for compounds 11, 12 and 13 were generated by Marvin Sketch and Tautomer Generator (Chemaxon Ltd., Budapest, Hungary) [47], which is freely accessible with academic license.
For kinetic aqueous solubility studies each sample was dissolved in DMSO to make 10 mM stock solutions. In a 96-well polypropylene plate (Greiner Bio-One, Kremsmünster, Austria), 15 µL stock solutions were added to 285 µL PBS (pH 7.4) to make starting donor solution with 500 µM as target concentration. For each sample, 3 replicates were measured. The samples were covered and shaken at 37 • C, 300 rpm for 2 h (Heidolph Titramax 1000, Heidolph Instruments GmbH & Co. KG, Schwabach, Germany). After that, each sample was transferred into a filter plate (MSSBLPC, Multiscreen Filter plate, Merck kGaA) and filtered (Vaccum Manifold, Merck kGaA, Darmstadt, Germany). The filtrates were transferred into HPLC vials, and acetonitrile (AcN) was added to aliquot to avoid precipitation. The final solvent ratio was AcN:PBS (70:30). Filtrate concentration was determined by HPLC-UV (see Section 2.1) using 3-points calibration.
Blood-brain barrier-specific (BBB) permeability measurements were carried out us-ing the PAMPA-BBB model. First, solutions with 500 µM target concentration were prepared as described for the kinetic solubility study. The solutions were sonicated for 10 min at room temperature. To prepare the artificial BBB-specific membrane, 16 mg BPLE were dissolved in 600 µL of n-dodecane:n-hexane (25:75). Each well of the donor plate (MultiscreenTM-IP, MAIPNTR10, pore size 0.45 mm, Merck kGaA) was coated with 5 µL lipid solution and fitted into the acceptor plate containing 300 µL PBS (pH 7.4) with 5% DMSO, and 150-150 µL of the PBS solutions (made from the DMSO stock solutions) were placed on the donor plate's artificial membrane. The sandwich plate system was covered with a tissue of wet paper and a plastic lid to avoid evaporation of the solvent, and it was incubated at 37 • C for 4 h. In the end, the initial 500 µM solutions (c D (0)), the donor (c D (t)) and acceptor solution (c A (t)) were analysed by HPLC-UV (see Section 2.1). BBB permeability was calculated using the effective permeability equation used for iso-pH conditions described by A. Avdeef [48] as follows.

Chemistry
In this work, a commercially available ginger extract was utilised to obtain significant amounts of our selected starting material, 6-gingerol (1), which could be obtained in a single-step purification using flash chromatography. Using compound 1 as a starting material, 6-Shogaol (2) was synthesised according to a literature method [43], and subsequent hydrogenation resulted in 6-paradol (3) [43] in an excellent yield. The reaction of 3 with hydroxylamine hydrochloride resulted in an oxime derivative as a mixture of (E/Z) isomers (4).
Reduction of the keto function of 6-gingerol resulted in 6-gingerdiol epimers 5 and 6 in a ratio of 3 to 2, respectively. The compounds were separated by preparative HPLC, and their structures were confirmed by HRMS and one-and two-dimensional NMR techniques, and by comparing their relevant spectral data to literature values [49][50][51]. Heating and cooling did not affect the stereoselectivity of the reaction. Compounds prepared directly from 6-gingerol are presented in Scheme 1. We also aimed to expand our study towards the chemical space around 6-dehydrogingerdione. To achieve this, a total synthetic approach was adopted (Scheme 2). First, 6-dehydrogingerdione (11) and its analogue lacking the aromatic methoxy function were synthesised starting from vanillin (7) or 4-hydroxybenzaldehyde (8) and 9-nonanedione according to a published method [44]. Catalytic hydrogenation of the double bond yielded We also aimed to expand our study towards the chemical space around 6-dehydrogingerdione. To achieve this, a total synthetic approach was adopted (Scheme 2). First, 6-dehydrogingerdione (11) and its analogue lacking the aromatic methoxy function were synthesised starting from vanillin (7) or 4-hydroxybenzaldehyde (8) and 9-nonanedione according to a published method [44]. Catalytic hydrogenation of the double bond yielded compound 13 (6gingerdione) [44]. The structure of compound 12 was confirmed using HRMS and NMR, and although the proton resonance of the enol methylene was not observed, the presence of carbon atoms at δ C 201.6 and 179.6 clearly showed that 12 undergoes keto-enol tautomerisation in solution. To investigate bioactivity changes upon replacing the keto group of compound 3 with an amide, we also prepared a capsaicin-like derivative (22). This was synthesised via a two-step procedure, starting with the catalytic hydrogenation of ferulic acid (19) to compound 20, followed by its DCC/DMAP-mediated coupling with heptylamine (Scheme 3).

Predicted and Experimental Physicochemical and BBB Penetration-Related Characterisation
The characterisation of the 6-gingerol derivatives was started with the determination of their physicochemical and blood-brain barrier (BBB)-specific permeability properties, which are commonly used in early-stage drug discovery. The study was carried out on two levels: (a) an in silico approach, using lead optimisation parameters and the Central Nervous System Multiparameter Optimisation (CNS MPO) compliance introduced by Wager et al. [52]; and (b) experimentally, through determination of kinetic solubility and in vitro BBB permeability of the test compounds. Predicted and experimental data are shown in Table 1. To facilitate exploring structure-activity relationships, oxygen and nitrogen-containing heterocyclic analogues were also prepared. Interestingly, the reaction of dehydrogingerdiol (11) and (12) with hydrazine monohydrate yielded two detectable products in both cases. Regarding the derivatives of 12, both products (15 and 16) were successfully isolated. On the other hand, in the case of compound 11, only compound 14 was isolated in a reasonable amount. Reactions of 11 and 12 with hydroxylamine hydrochloride resulted in oxazines 17 and 18, respectively. Unfortunately, some of the non-protonated aromatic carbons were not detected in the 13 C NMR spectrum due to their long relaxation time after the irradiation, but the characteristic 1 H proton resonances, HRMS spectra and spectral behaviour of the compounds allowed the establishment of their chemical structures.
To investigate bioactivity changes upon replacing the keto group of compound 3 with an amide, we also prepared a capsaicin-like derivative (22). This was synthesised via a twostep procedure, starting with the catalytic hydrogenation of ferulic acid (19) to compound 20, followed by its DCC/DMAP-mediated coupling with heptylamine (Scheme 3). To investigate bioactivity changes upon replacing the keto group of compound 3 with an amide, we also prepared a capsaicin-like derivative (22). This was synthesised via a two-step procedure, starting with the catalytic hydrogenation of ferulic acid (19) to compound 20, followed by its DCC/DMAP-mediated coupling with heptylamine (Scheme 3).

Predicted and Experimental Physicochemical and BBB Penetration-Related Characterisation
The characterisation of the 6-gingerol derivatives was started with the determination of their physicochemical and blood-brain barrier (BBB)-specific permeability properties, which are commonly used in early-stage drug discovery. The study was carried out on two levels: (a) an in silico approach, using lead optimisation parameters and the Central Nervous System Multiparameter Optimisation (CNS MPO) compliance introduced by Scheme 3. Preparation of the amide analogue of 6-paradol (22). Reaction conditions: a. H 2 /Pd/C/ EtOAc/rt; b. DCC/DMAP/CH 2 Cl 2 /rt.

Predicted and Experimental Physicochemical and BBB Penetration-Related Characterisation
The characterisation of the 6-gingerol derivatives was started with the determination of their physicochemical and blood-brain barrier (BBB)-specific permeability properties, which are commonly used in early-stage drug discovery. The study was carried out on two levels: (a) an in silico approach, using lead optimisation parameters and the Central Nervous System Multiparameter Optimisation (CNS MPO) compliance introduced by Wager et al. [52]; and (b) experimentally, through determination of kinetic solubility and in vitro BBB permeability of the test compounds. Predicted and experimental data are shown in Table 1.
Combining Lipinski's rule of five (Ro5) [53] and CNS MPO [52] criteria systems, the proton dissociation and lipophilicity properties of 6-gingerol derivatives were compared in a first approach. Regarding their acid-base character, the compounds can be classified into three groups: (i) monoprotic phenols (1-6, 17, 18 and 22), (ii) diprotic amphoterics (imidazole derivatives; 14-16), and (iii) diprotic acids (gingerdione derivatives; 11-13). The proton-dissociation behaviour of gingerdione derivatives is particularly interesting due to their tautomeric states shown in Scheme 3. The distribution % of the A-C tautomeric states of derivatives 11-13 was generated using Chemaxon Ltd.'s freely available Tautomer Generator plugin. This suggests compounds 11-13 to be mostly present in their enol forms (A:B, ∼30:60%), while the dione forms with C-H acid function are much less expressed (C, ∼10%). It is also important to note that, while in the case of the other gingerols (1-6, 14-22) the strongest pK a,acid values refer to the aromatic OH function, the pK a,acid parameters indicate the proton-dissociation behaviour of the enol and dione C-H acid functions for 11-13. Thus, based on these pK a,acid values, two conclusions can be made. First, in the case of 11-13, the stronger acid function can be assigned to the enol and C-H acid moiety, and second, the acidic character of the A, B tautomeric forms is stronger than that of the C forms. In terms of lipophilicity, all 6-gingerol derivatives meet the drug-likeness criterion of Ro5 (logP < 5). At the same time, based on the two-level risk classification created by the leadlike [54] and CNS MPO criteria (logD pH7.4 ), 2-4 and 14-18 exceed (see Table 1, magenta: high violation), and 22 approach (yellow: moderate violation) the logD pH7.4 violation limit.
The lipophilicity values of the tautomers of 11-13 also show a marked difference. For all three compounds, the enol forms A/B (11A, 12A, 13B) are more lipophilic than the dione forms (C), which also manifests in the corresponding CNS MPO values. Regarding the next medchem parameter, the polar surface area (TPSA), all the tested gingerols satisfy both the Veber's rule (30Å 2 < TPSA < 140Å 2 ) for bioavailability [55] and the CNS MPO range (40Å 2 < TPSA < 90Å 2 ) [52]. CNS MPO values (physchem-based CNS compliance) of gingerols were evaluated by a three-level classification system (green-yellow-magenta) for easier overview. Compounds 1-3, 5-6, 11-13, 17 and 22 (green-yellow) can be considered as suitable candidates for further CNS-targeted preclinical studies. In the case of 11-13, due to the lipophilicity and HBD differences of the tautomers, the C tautomeric forms carry the optimal CNS MPO character. In addition to the predicted parameters, experimental kinetic solubility (PBS, pH 7.4) and in vitro BBB-specific permeability (PAMPA-BBB) of the compounds were also determined. Also applying a three-level classification for the kinetic solubility values, the compounds 1, 2, 5, 6, 13-15 and 22 were in the acceptable range (greater than 100 µM). In parallel, due to limitations resulting from poor solubility, we could only determine the permeability of these gingerols using PAMPA-BBB study. In Table 1, compounds 1, 2, 5, 6 and 22 are highlighted in green (P e,BBB ≥ 25·10 −7 cm/s), for which we identified increased BBB permeability. In addition to these compounds, 14-15 are also adequate for BBB penetration. In the case of 13, the increased hydrophilic character may impair the BBB permeability.

Antiplatelet Aggregation and COX-1 Inhibition Activity of the Compounds
The arachidonic acid (AA)-cyclooxygenase-1 (COX-1) pathway plays an important role in platelet activation [56]. Aspirin, the standard antiplatelet drug, can prevent AA metabolism to thromboxane A2 by inhibiting COX-1, and thus exert antiplatelet effects [57]. To evaluate similar bioactivities of 6-gingerol (1) and its derivatives, the compounds were tested for their inhibitory effects on AA-induced platelet aggregation . Compounds 3, 2,  17, 16, and 13 showed the most promising results, with IC 50 values of around 2-4 µM, respectively, while the effects of 6-gingerol (1) and aspirin were up to 22 and 50 times weaker, respectively (Table 2). Table 2. Antiplatelet and Cyclooxygenase-1 (COX-1) inhibition assay results. For antiplatelet activity, washed human platelets were treated with gingerol derivatives for 3 min and then stimulated with arachidonic acid (100 µM); data are presented as mean ± SEM, n = 3 for antiplatelet and n = 2 for COX-1 inhibition assay; aspirin was used as a positive control.

Compound
Antiplatelet IC 50  (LLE (Antiplatelet) ≥ 1.5 or LLE (COX-1) ≥ 1.0 and IC 50 ≤ 10 µM). b For the sake of strict characterisation, the tautomer with the highest logP value (see Table 1) was included in the LLE calculation for compounds 11-13, i.e., the value for the worst possible case is shown in the table.
The antiplatelet mechanism of action of 6-gingerol and its derivatives has previously been suggested to be COX-1 inhibition [37,58,59]. The results obtained for the newly synthesised compounds are in accordance with this notion. The IC 50 datasets of Table 2 give a linear correlation coefficient (R 2 ) value of 0.887 (Supplementary Materials, Figure S26), strongly suggesting that COX-1 inhibition is indeed the mechanism behind the observed antiplatelet action.
Concerning structure-activity relationships, our results suggest the importance of an aromatic methoxy group, as seen by comparing the IC 50 values of 11 vs. 12, 14 vs. 15, and 17 vs. 18. Interestingly, this rule did not apply for compound 16. This may underline a possible role of the ∆ 1,2 olefin in some cases, e.g., when it is conjugated with a pyrazole ring. Presence of the 5-OH group is highly unfavourable; its elimination (as in compounds 2, 3, and 4) increased antiplatelet activity by ca. an order of magnitude, as well as its oxidation (compound 13 vs. 1) or replacement by a heterocycle (e.g., compound 14 vs. 1).
Compound 22, in which the β-keto alcohol function of compound 1 was replaced by an amide group, showed only moderate activity. Notably, however, replacing the 3oxo group by an oxime group led to only a slight, ca. 2-fold decrease in the antiplatelet activity (IC 50 = 5.2 µM), while this chemical change opened the way to diverse further functionalisation possibilities.
Our results come in accordance with previous reports on compound 3, i.e., 6-paradol that is naturally present in ginger roots.
Since the mid-2000s, the entropic or lipophilicity-driven lead optimisation constraint has mainly been observed in CNS-targeted drug discovery, significantly increasing the number of clinical candidates that were promiscuous or otherwise carrying off-target effects and toxicity risks [60]. To reduce this issue, several ligand efficiency metrics have been introduced [61], which can be used to filter out this effect. In this context, in our study, the IC 50 data obtained on the two biological targets were evaluated using the ligand-lipophilic efficiency (LLE = pIC 50 −logP) metric [60]. After, a two-conditions-based lead selection was performed using the IC 50 and LLE values, where the LLE values of 6-gingerol derivatives with an IC 50 ≤ 10 µM and an LLE value higher than 1.0 were highlighted ( Table 2, compounds coloured green for LLE Antipatelet : 2, 3, 11, 13 and 17 and LLE COX-1 : 3). Summarizing the data in Tables 1 and 2, due to the strict selection, only compound 2 can be assigned as a primary candidate for further preclinical studies, while compounds 3, 11 and 17 are potential leads that require an appropriate formulation to improve their aqueous solubility. From the point of view of lead optimisation, compound 3 is particularly interesting, satisfying the LLE/IC 50 screening criteria for both biological targets. Compound 13 can also be identified as a secondary lead, for which the further goal may be fine-tuning the BBB permeability property.

Molecular Docking
Compounds were docked using AutoDock4 into the human COX-1 enzyme crystal structure retrieved from the Protein Data Bank (PDB ID: 6Y3C). Grid parameters were set to centre at residue Ser530 and to include residues Tyr385, Arg120, and Tyr348 that are in the COX-1 binding pocket [62]. In the cases of compounds 11, 12, and 13, the tautomeric forms A-C were subjected to in silico docking. Detailed results of the docking study are provided as Supplementary Materials , Table S1.
The highest binding affinity (−9.5 Kcal/mol) was found for the new isoxazole-containing compound 17, which was also among the most potent antiplatelet derivatives. L-Toth et al. discussed the importance of Tyr385 and Ser530 in the irreversible binding of aspirin to COX-1 active site; notably, the docking results showed hydrogen bonding interactions with Ser530 in the case of compound 1, 4, and 6, while compounds 11B, 12A, and 22 appeared to interact via hydrogen bonds with Tyr385 (Table S1). Interestingly, compound 17 did not show any of the above-mentioned interaction with either of these amino acids (Figure 1).

Antioxidant Assay
Antioxidant activity was assessed using multiple models, including DDPH, ORAC, ONOO − scavenging, and XO inhibition assays; results are shown in Table 3. Compounds 5, 17, 4, 1, and 11 showed the best activity in the diphenyl-2-picrylhydrazyl (DPPH) scavenging capacity assay. Among these, compound 5 is clearly the most promising antioxidant lead, considering its DPPH-scavenging IC 50 value, its ORAC value, which is more than twice as potent as that of Trolox, and its predicted and experimental pharmacokinetic parameters. Interestingly, compound 6, the three-epimer variant of 5, showed only ca. half of the activity of 5 in the DPPH assay and was also weaker in terms of ORAC. Similar to the results obtained for the antiplatelet and COX-1 inhibition assay, the importance of an aromatic methoxy group for potent DPPH scavenging activity was highlighted; except for compound 16, all compounds without this moiety (12, 15, and 18) were inactive in this regard. Concerning their ORAC values, however, compounds 12 and 16 were the most potent among all compounds, which highlights the complementary value of these two bioassays to evaluate free radical scavenging activity of small molecule antioxidants. aspirin to COX-1 active site; notably, the docking results showed hydrogen bonding interactions with Ser530 in the case of compound 1, 4, and 6, while compounds 11B, 12A, and 22 appeared to interact via hydrogen bonds with Tyr385 (Table S1). Interestingly, compound 17 did not show any of the above-mentioned interaction with either of these amino acids (Figure 1).

Antioxidant Assay
Antioxidant activity was assessed using multiple models, including DDPH, ORAC, ONOO − scavenging, and XO inhibition assays; results are shown in Table 3. Compounds 5, 17, 4, 1, and 11 showed the best activity in the diphenyl-2-picrylhydrazyl (DPPH) scavenging capacity assay. Among these, compound 5 is clearly the most promising antioxidant lead, considering its DPPH-scavenging IC50 value, its ORAC value, which is more than twice as potent as that of Trolox, and its predicted and experimental pharmacokinetic parameters. Interestingly, compound 6, the three-epimer variant of 5, showed only ca. half of the activity of 5 in the DPPH assay and was also weaker in terms of ORAC. Similar to the results obtained for the antiplatelet and COX-1 inhibition assay, the importance of an aromatic methoxy group for potent DPPH scavenging activity was highlighted; except for compound 16, all compounds without this moiety (12, 15, and 18) were inactive in this regard. Concerning their ORAC values, however, compounds 12 and 16 were the most potent among all compounds, which highlights the complementary value of these two bioassays to evaluate free radical scavenging activity of small molecule antioxidants.  Table 3. The antioxidant activity results of 6-gingerol and its derivatives. Values are given as mean ± standard error of the mean (SEM). ORAC assay results are given in Trolox equivalents (TE), ONOO − scavenging and xanthine oxidase (XO) inhibition assay results are given in % inhibition at concentrations of 500 and 100 µM, respectively; n = 2 for DPPH; n = 3 for ORAC, ONOO − and XO.

Conclusions
In this study, a combination semi and total synthetic strategy was adopted to prepare fourteen 6-gingerol derivatives including eight new compounds, which were subsequently characterised as antiplatelet and COX-1 inhibitor agents, and free radical scavenger or XO inhibitor antioxidants. The compounds' pharmacodynamic and pharmacokinetic characterisation revealed 6-shogaol (2) to be the best lead as a cardiovascular protective agent, and compounds 3, 11, and 17 as new starting points for hit-to-lead optimisation. The 3,5-diol compound 5 was identified as a more potent and less promiscuous antioxidant than its parent compound 6-gingerol (1) or its three-epimer compound 6. Due to its favourable pharmacokinetic parameters, compound 5 is also suggested as a potential CNS-specific antioxidant. Further studies to evaluate this notion are to be conducted soon.