In Silico Pharmacokinetics Study of 2,5-Dibenzylidenecyclopentanone Analogs as Mono-Ketone Versions of Curcumin

. The absorption-distribution-metabolism-excretion (ADME) profile is a crucial parameter that indicates the pharmacokinetics of the drug. The pharmacokinetic properties of a drug represent the fate of the drug in the body. Curcumin is a main compound in turmeric produced by plants of the Curcuma longa species, and has several pharmacological effects in animal and human clinical studies. However, preclinical and clinical studies have shown that curcumin has pharmacokinetic limitations such as poor bioavailability and rapid metabolism which restrict its widespread use. Therefore, various modifications and synthesis of some analogs using curcumin as a lead compound with variations in the main structure and attached substituents have been carried out to explore the pharmacological effects as drug candidates. One of the widely developed methods is the modification of curcumin's main structure, specifically the conversion from diketone to mono-ketone.In 1997, 2,5-dibenzylidene cyclopentanone analogs were synthesized and their biological activity were performed. However, there is no further information related their pharmacokinetic properties. Therefore, those properties were predicted by performing ADME calculation in two online servers, ADMETsar 2.0 and ADMETlab 2.0.. By utilizing the online servers ADMETsar 2.0, and ADMETLab 2.0 for in-silico screening of pharmacokinetic properties, from the 17 compounds, it was found that the variation among pharmacokinetic aspects was observed, either decreasing or increasing drug likeness properties of 2,5-dibenzylidene cyclopentanone analogs compared to curcumin. In addition, the interaction those analogs with protein or enzymes involved during ADME process such as blood plasma protein (albumin), p-Glycoprotein, and CYP3A4 was evaluated by performing molecular docking.. The docking results showed a sufficiently positive correlation with ADME screening outcomes


INTRODUCTION
The inability of a drug to enter the market is often associated with issues in the testing of drug efficacy and safety.The safety and efficacy of a drug are influenced by its absorption, distribution, metabolism, excretion (ADME), as well as its side effects/toxicity (T).Typically, methods used to assess the ADME/T properties of a drug involve animal testing (Phase I Clinical Trials), which can be costly and time-consuming.Phase I Clinical Trials also raise ethical concerns (animal rights) and economic *Corresponding Author : nugroho_ae@ugm.ac.id considerations.One potential solution is the application of rational drug design methods using computational approaches [1].
Currently, a significant number of QSAR/SAR models have been developed, and specialized programs available on online servers have been created and developed based on these models.The advantages of these methods are the ability to accurately predict the ADME/T properties of a chemical compound and enable screening of most candidate compounds or their properties before conducting in vivo and in vitro testing.This approach helps to save resources, costs, and test animals [2].
Curcumin [1,7-bis-(4-hydroxy-3-methoxyphenyl ) -1,6-heptadiene-3,5-dione] (figure 1) is a compound isolated from Curcuma longa L. It has been widely known and used for centuries as a food pigment and spice.Curcumin has been found to have various traditional pharmaceutical applications in diseases, including external/internal wounds, liver diseases (especially jaundice), blood purification, antimicrobial effects, and inflammation in joints.In modern drug research, curcumin is still considered a highly promising compound for drug design and development based on its explicit bioactivity, minimal toxicity, and ease of synthesis.Curcumin [1,7-bis-(4-hydroxy-3methoxyphenyl)-1,6-heptadiene-3,5-dione] However, preclinical and clinical studies have shown that curcumin has pharmacokinetic limitations such as poor bioavailability, rapid metabolism, and the need for repeated oral doses, which restrict its widespread use.Curcumin is stable at pH below 6.5.Its instability above pH 6.5 is caused by the methylene group.By removing the methylene group and one carbonyl group, B.M. Markverich, M. Artico, and H.I. El-Subbagh synthesized a series of mono-carbonyl analogs of curcumin, 1,5diaryl-1,4-pentadien-3-one, and evaluated their bioactivity.The results showed that the mono-carbonyl analogs exhibited stronger inhibitory effects in various cancer cells compared to curcumin [3].
One synthetic analog compound with a structure similar to curcumin, containing a mono-carbonyl group, is 2,5-Di2,5-dibenzylidenecyclopentanone, which was synthesized by Sardjiman in 1997 as part of a series of 17 analog compounds.The research conducted by Sardjiman in 1997 produced analog compounds of 2,5dibenzylidenecyclopentanone with variations in substituents at carbon atoms 4, 5, and 6 that are symmetric in both benzylidene structures.In this study, a pharmacokinetic evaluation was performed by conducting ADME profiling of the analog compounds of 2,5dibenzylidenecyclopentanone using a virtual screening approach with online servers ADMETLab 2.0, and ADMETsar 2.0 [4] [5] .

Instrumentation
Data processing was performed using a PC device with the following specifications: i5 processor, 8 GB RAM, 512 GB SSD, and Windows 10 operating system.

Data Source of Compounds
The studied compounds were the curcumin and 17 analog compounds of 2,5-dibenzylidenecyclopentanone that were synthesized and investigated by Sardjiman in 1997.The data on the structure of the analog compounds of 2,5dibenzylidenecyclopentanone consisted of a table containing a list of 17 derivative compounds of curcumin that were synthesized in Sardjiman's research in 1997 (Figure 2) (table 1) [6].The list of compounds represents a part of the 2,5-dibenzylidenecyclopentanone analogs with variations in functional groups attached to carbon atoms 3, 4, and 5 on the benzene ring attached to both sides of the cyclopentanone structure.

DATA PROCESSING
The compound structures were inputted into online servers that provide ADME profile screening, namely ADMETsar 2.0 and ADMETlab 2.0.The online servers were accessed through a web browser by visiting the screening pages at https://admetmesh.scbdd.com/, and http://lmmd.ecust.edu.cn/admetsar2/ .Once the online servers were opened, the compound structures were entered into the provided input columns one by one.
The input of compound structures into the online servers for screening can be done in two ways: directly drawing the compound structure using the available tools or entering the SMILES (The simplified molecular-input line-entry system) code, which is a chemical notation that represents the molecular structure as a graph with optional chiral indications.After inputting the structure, the "submit" or "Run!" option was selected to start the screening process.If a manually drawn structure was entered, the SMILES code would appear first.
After the screening of the entered compounds, ADME/T data along with the physical-chemical properties of the compounds would be obtained.The screening results would produce positive/negative signs, probability values, or absolute numbers according to the parameters, indicating the capability or lack thereof, as well as the degree, of the compound's ability to undergo ADME processes in the body, based on the respective database used by each online server.The subsequent analysis involved interpreting and comparing the physicalchemical properties and ADME profiles of the same compounds screened using three different online servers and among the compounds in the analog list.

Data Analysis
The obtained ADME property data were then analyzed for their drug-likeness properties using various pharmacokinetic parameters, including physical-chemical properties, adsorption capacity, distribution ability, metabolic capacity, and elimination ability.The focus of this research was to compare the ADME properties of curcumin obtained with those of the analog compound 2,5-dibenzylidenecyclopentanone as its mono ketone version.The online servers, ADMETlab 2.0 and ADMETsar 2.0, have similarities in data presentation, where the majority of the data is qualitative, both in terms of absolute value parameters and parameters that generate probability data.However, there are differences between them.ADMETlab 2.0 for probabilistic parameters shows values representing the compound's usefulness (specifically referring to drug likeness of compounds) using symbols such as "+++" for the highest range of values (probability 0.9 -1) and "---" for the lowest range of probability values (0-0.1).In this research, the probability values taken are the upper limits (endpoint probability) as the chosen values to indicate the level of probability.
In contrast, ADMETsar provides comprehensive values ranging from 0 to 1 to indicate the probability results of the inputted compounds.Due to these differences, the correlation between the two cannot be determined, but it merely indicates the presence of two prediction values from two different servers.The research was conducted by inputting the SMILES code of each tested compound into all three types of online servers, followed by pharmacokinetic analys    Expected decrease in pharmacokinetic properties (The pharmacokinetic characteristics of the analog 2,5-dibenzylidene cyclopentanone are already good, but not as good as curcumin) Similarity to curcumin (the pharmacokinetic characteristics of the analog 2,5-dibenzylidene cyclopentanone that are identical to curcumin) Unexpected increase in pharmacokinetic properties (the pharmacokinetic characteristics of 2,5-dibenzylidene cyclopentanone that are poor, but not as poor as curcumin) Unexpected decrease in pharmacokinetic properties (The pharmacokinetic characteristics of 2,5-dibenzylidene cyclopentanone that are poor and worse than curcumin)    The screening results of good physicochemical properties using the ADMETsar 2.0 (table 2), and ADMETLab 2.0 (table 3) servers show that the analogs of 2,5-Dibenzylidene cyclopentanone tend to have better physicochemical properties compared to curcumin, except for the hydrogen bond acceptor and donor parameters.All analogs of 2,5-Dibenzylidene cyclopentanone exhibit lower polar surface area compared to curcumin, except for compound B15, where the variation in its substitution is two methoxy groups on each side, which is twice the amount compared to curcumin.The addition of oxygen atoms is strongly suspected as the reason for the increase in the polar surface area value.
Both curcumin and all analogs of 2,5-Dibenzylidene cyclopentanone almost meet the RO5 rules, with deviations not exceeding 1, mostly due to the log P value exceeding 5 in 11 compounds of 2,5dibenzylidenecyclopentanone analog.Among all the tested compounds, only curcumin and six analogs of 2,5dibenzylidene cyclopentanone meet the criteria of having a logP value below 5.This is because of the presence of substituents that increase the non-polar properties such as C, F, and Cl, without being balanced by the presence of O.However, there is one compound that deviates from the 2 rules, making it not pass the RO5, namely compound B14.This is due to its substitution of C4H9 (figure 3), which is a relatively long hydrocarbon chain, resulting in a total molecular weight exceeding 500 g/mol and polarity value of more than 5.

Absorption
Based on the screening results using ADMETLab 2.0, it was found that the ability of the 2,5-dibenzylidene cyclopentanone analogs to serve as P-Glycoprotein substrates is better than curcumin (table 2).
Conversely, the inhibitory effect on P-Glycoprotein is lower for all of compounds of 2,5dibenzylidene cyclopentanone analogs compared to curcumin.The results of the prediction using ADMETlab 2.0 indicate that the tendency of the analogs to be Pglycoprotein inhibitors is higher than that of substrates, compared to curcumin.However, different from the screening results using ADMETsar 2.0, although the ability as a substrate is very high with probabilities close to 1 and better than curcumin, the inhibitory ability of the 2,5-dibenzylidene cyclopentanone analogs against P-Glycoprotein is almost always higher (table 3).
The difference in screening results for ligandprotein interactions regarding the parameter of interaction with P-glycoprotein between the two servers cannot be explained at the physicochemical level.This is because such interactions are specific ligand-receptor interactions that require further modeling and in vivo or in vitro research to validate them.However, both ADMETLab 2.0 and ADMETsar 2.0 servers show a high tendency for binding as substrates to P-Glycoprotein.This can be interpreted positively as the ability to enter cells to achieve therapeutic effects.However, excessive binding as a substrate can lead to negative effects, such as resistance to the efflux of substances from cells.P-gp is a transporter protein located on the cell membrane and plays a role in removing foreign compounds or drugs from within the cell.By strongly binding to P-gp, the compound can inhibit the normal function of P-gp in removing drugs from within the cell, leading to an increase in the drug concentration inside the cell and reducing its effectiveness.As a result, the compound can cause drug resistance, especially in cancer treatment cases.
Chemotherapy resistance is often associated with high Pgp activity, which can result in a decrease in the concentration of cytotoxic drugs inside cancer cells and reduce the effectiveness of treatment.Additionally, increased interaction with P-Glycoprotein can also affect the systemic bioavailability of the drug, potentially reducing the desired therapeutic effects.The docking simulation results reveal interactions between curcumin and analogs of 2,5dibenzylidene cyclopentanone, consistently demonstrating interactions with amino acid residues in the p-glycoprotein transport channel (figure 4).This indicates a favorable predictive affinity for both curcumin and the analog compounds of 2,5-dibenzylidene cyclopentanone.
In terms of the Human Intestinal Adsorption (HIA) parameter, the screening results using ADMETLab 2.0 showed that almost all of the 2,5-dibenzylidene cyclopentanone analogs have similarly poor intestinal absorption as curcumin, approaching 0, except for compounds B13 and B14, which are compounds with long hydrocarbon chain substituents (C3H7 and C4H9).This is thought to be due to an increase in solubility in the lipid layer of the intestinal membrane (increased non-polar properties that enhance solubility in the non-polar lipid phase).With these substituents, the solubility in the lipid phase in the intestine is increased, resulting in a higher ability to reach systemic circulation.Low HIA ability is associated with a large molecular size, which hinders penetration through the intestinal absorption barrier.Additionally, the presence of OH makes the compound sufficiently polar, reducing its solubility in the lipid layer of the intestinal membrane.However, the opposite prediction is shown by the ADMETsar 2.0 server, where all test compounds, including curcumin, are displayed as having very high intestinal absorption with probabilities close to 1.However, this prediction is less reliable considering various experimental studies that have shown very low intestinal absorption of curcumin and its derivatives.

Distribution
The distribution parameters examined in this study include the volume of distribution (VD), binding to Plasma Protein Binding (PPB), and blood-brain barrier (BBB) penetration ability.Based on the screening results with ADMETLab 2.0 (table 2), the VD values of the 2,5dibenzylidenecyclopentanone analogs are generally higher than that of curcumin, except for compound B2, which has an OH group as a substituent.This may be due to the polar group experiencing an increase in polarity after passing through the metabolism phase.The OH group also has a strong binding ability to plasma proteins, reducing its distribution capability in systemic circulation.Additionally, the polar OH group can form hydrogen bonds with water, limiting its solubility to aqueous compartments such as blood and extracellular fluids, which have a relatively smaller volume compared to adipose tissues.The highest and most significant VD value is found in compound B14, which has a long hydrocarbon chain substituent (C4H9).This is likely due to its non-polar nature, which increases its solubility in adipose tissues, the largest distribution compartment in the body.In the docking simulation, an affinity is observed between curcumin and analogs of 2,5-dibenzylidene cyclopentanone, indicating specific interactions with the long-chain hydrocarbon substituent (B14) and His and Lys residues that form hydrophobic interactions with the hydroxyl group.This suggests a strong binding between these compounds (as shown in Figure 5 displaying PDB interactions with albumin) with a docking score representing a Gibbs free energy of -8.556 kcal/mol.Such interactions are not seen with curcumin, which demonstrates weak interactions with albumin.
Consistent with the theory, contrary to the value of VD, the screening results with ADMETLab 2.0 show that the binding ability of the test compounds to plasma proteins, both curcumin and analog 2,5-dibenzylidene cyclopentanone, is moderate to high.
Based on the theory in the equation of volume of distribution that uses the fraction of drug in the body as the multiplying factor and the fraction of free drug in the systemic circulation as the dividing factor, it can be concluded that drugs with strong binding to plasma proteins have low volume of distribution values.This is because the higher the binding between a compound and plasma proteins, the higher its fraction in the free form in the systemic circulation, resulting in lower volume of distribution values.This is consistent with the screening results from ADMETLab 2.0, where although all compounds fall within the optimal VD range (0.04-20 L/kg), these values are much closer to the lower limit than to the upper limit of the optimal VD range, in contrast to the close to 100% overall protein plasma binding.The strong binding of drugs to plasma proteins is often considered unfavorable as it reduces the drug's ability to distribute freely in the systemic circulation and reach its target.Drugs that bind strongly to plasma proteins tend to remain in a bound form and cannot easily transfer to target tissues, thereby potentially affecting the effectiveness of therapy.Drugs that bind strongly to plasma proteins are more likely to concentrate in the blood and other aqueous compartments, while distribution to non-aqueous tissues such as fat tissue or specific organs may be limited.Furthermore, if a drug is tightly bound to plasma proteins and difficult to release, it may have a longer half-life, leading to an increased risk of drug accumulation and undesirable side effects.However, for specific purposes, the binding ability to plasma proteins can be beneficial, such as helping to maintain a more stable therapeutic drug concentration in the blood, reducing the risk of rapid elimination through kidney filtration, and protecting the drug from rapid breakdown or elimination.
Based on the screening results with ADMETLab 2.0, the prediction of blood-brain barrier (BBB) penetration ability shows very low probabilities ranging from 0 to 0.3 (table 2), with a tendency for lower penetration compared to curcumin.In contrast, the prediction with ADMETsar 2.0 (table 3) shows that almost all compounds have BBB penetration probabilities above 0.5.The ability of a drug to penetrate the BBB is not always a priority.Some drugs indeed need to penetrate the BBB to reach therapeutic targets in the brain, such as in the treatment of neurological disorders or mental disorders.Examples include drugs used in the treatment of epilepsy, depression, or Alzheimer's disease.For these drugs, the ability to cross the BBB becomes an important criterion for therapeutic effectiveness.However, not all drugs need to penetrate the BBB.Many drugs are designed to affect organs or tissues outside the brain, such as the heart, lungs, or other organs in the body.For these drugs, the BBB acts as a desired barrier, protecting the brain from side effects or potential damage caused by the drugs.
The physicochemical factors that influence the ability of a compound to penetrate the Blood-Brain Barrier (BBB) are molecular size, polarity, and lipophilicity.Compounds with small molecular size, non-polar (lipophilic) nature, and the ability to easily cross cell membranes tend to have better BBB penetration.On the other hand, compounds with large molecular size, polar nature, and low lipophilicity are more likely to face difficulties in crossing the BBB due to the barrier formed by the brain endothelial cells, which restricts the penetration of polar and large molecules from the bloodstream to the brain.Due to the majority of analogs of 2,5-Dibenzylidene Cyclopentanone being compounds with low polarity (logP>5), theoretically, the screening results would indicate that these compounds are more likely to easily penetrate the Blood-Brain Barrier (BBB).Therefore, based on these physicochemical properties, the predictive results from ADMETsar 2.0, which show high BBB penetration probability (probabilities>0.5),are more accurate compared to ADMETLab 2.0, which indicates probabilities not exceeding 0.3 for all test compounds.

Metabolism
Metabolism is the process of chemical modification of a drug within the body.It refers to the biotransformation of compounds in drugs so that they can be eliminated more easily.Most drug metabolism processes occur in the liver, as the enzymes that facilitate drug metabolism reactions are concentrated there [9].
Phase I metabolism is a series of reactions where functional groups on the drug are exposed for the first time, aiming to increase the compound's polarity.Phase I metabolism occurs in various tissues, with hepatic circulation being the main pathway.Other tissues involved include gastrointestinal epithelium, kidneys, skin, and lungs.Within the cells, most phase I enzymes are located in the endoplasmic reticulum, which is rich in microsomes [9] [10].The major catalyst in phase I metabolism reactions is the cytochrome P450 system (figure 6), a large family of membrane-bound enzymes found in the endoplasmic reticulum of hepatocytes.These enzymes are regulated by constitutive (relatively constant abundance regardless of substrate presence) and inducible (activated under specific conditions) regulation.The same gene sequence is used to classify cytochrome P450 enzymes, using family numbers (numeric notation) and subfamily letters (alphabetic notation) to distinguish between isoforms or individual enzymes.Subtypes of cytochrome P450 include CYP1A1/2, CYP1B1, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1, CYP3A4/5/7.The most common metabolic pathways involve CYP3A4/5, CYP2C9, CYP2D6, and CYP2C19.CYP3A4 is the most abundant isoform in the human liver and contributes to about 50% of all CYP450 activities [9] [11].
In the screening conducted, the ability of test compounds to bind as substrates to P450 is considered a desired pharmacokinetic property because the prediction indicates the likelihood of the compounds being metabolized and eliminated.Conversely, the ability to inhibit P450 is considered an undesired property because P450 inhibition can lead to various drug interaction issues, reduced drug effectiveness, and drug accumulation with potentially toxic properties due to the decreased ability of P450 to metabolize various types of drugs and other compounds [12].
Both the substrate and inhibitor abilities cannot be determined solely based on physicochemical properties.This is because the enzyme-compound complex is a ligand-receptor interaction that is specific to the chemical and stereochemical interactions of the compound with the active site of the enzyme [13].Some screening results also indicate that the tested compounds can act as both inhibitors and substrates.This is strongly suspected to be related to the mechanism of competitive inhibition.However, further research is needed to determine the mechanism of action as a substrate or inhibitor.
The screening results with ADMETsar 2.0 and ADMETlab 2.0 show different results when considering each metabolic parameter, specifically isoforms of P450 (CYP1A1, CYP2C19, CYP2C9, CYP2D6, CYP3A4).For example, the screening results with ADMETlab 2.0 indicate that the ability of 2,5dibenzylidenecyclopentanone analogs as CYP2C9 substrates is generally good but not better than curcumin.On the other hand, the screening results with ADMETsar BIO Web of Conferences 75, 04002 (2023) https://doi.org/10.1051/bioconf/20237504002BioMIC 2023 2.0 indicate that 8 compounds have better CYP2C9 substrate abilities compared to curcumin.Another example is the prediction of CYP3A4 inhibition.The screening results with ADMETlab 2.0 indicate that 2,5-dibenzylidenecyclopentanone analogs tend to inhibit CYP3A4 but not are stronger than curcumin.However, the opposite results are found in the prediction results with ADMETsar 2.0, where almost all 2,5dibenzylidenecyclopentanone analogs have stronger inhibitory abilities than curcumin.
Based on their structures, the differences in substrate-inhibition potency can be studied for substituents attached to the aromatic ring of analogs of 2,5-dibenzylidene cyclopentanone [14].Isoform P450 specifically oxidizes aromatic groups through hydroxylation at the ortho, meta, and para positions of the aromatic ring (figure 7).Specifically, CYP1A1 and 1A2 tend to oxidize aromatic groups at the ortho and para positions.These isoforms are often involved in the metabolism of polycyclic aromatic compounds, such as benzo[a]pyrene and other polycyclic aromatics [15].CYP2C9 tends to oxidize aromatic groups at the para position, as demonstrated by its ability to oxidize warfarin at the para position of its aromatic ring [16].CYP2D6 tends to oxidize aromatic groups at the ortho and para positions.For example, CYP2D6 can oxidize substrates such as debrisoquine at the ortho and para positions of its aromatic ring [17].Among the isoforms of P450, CYP3A4 has the broadest range of oxidation indices due to its ability to hydroxylate at the ortho, meta, and para positions [18].Among the analogs of 2,5-dibenzylidene cyclopentanone, the compound most similar to curcumin is PGV-0 (Pentagamavunon-0) or in this study referred to as the compound with code B1.The similarity to curcumin is evident from the presence of one methoxy substituent at the meta position and one hydroxyl substituent at the para position.Theoretically, if there are no substituents on the aromatic positions of the aromatic ring and only hydrogen is present (as in the case of compound B2), its ability to act as a substrate would be higher, and its ability to act as an inhibitor would be lower compared to other compounds that have substituents on their aromatic rings.Screening results with ADMETLab 2.0 showed that compound B2, without any substituents on its aromatic ring, actually has a higher probability of inhibition compared to other compounds.In contrast, screening results using ADMETsar 2.0 showed that the inhibition ability of compound B2 is lower than other compounds.This indicates that the predictions for the metabolism parameter conducted by ADMETsar 2.0 are more accurate according to theory compared to ADMETLab 2.0.However, despite the differences between the two servers, there is one common observation: the tendency to be a substrate is lower than curcumin, and the inhibition ability is generally higher than curcumin.Curcumin is a natural compound, and natural compounds generally have high P450 inhibition potency.The screening results for analogs of 2,5-dibenzylidene cyclopentanone show unexpected outcomes, with a higher inhibition potency and lower substrate capability compared to curcumin.
Although there are differences between the two servers, there is a general trend of lower substrate ability compared to curcumin and higher inhibition ability compared to curcumin.Curcumin is a natural compound and natural compounds generally have relatively high P450 inhibition capabilities [19].The screening results for 2,5-dibenzylidenecyclopentanone analogs, on the other hand, show unexpected results with a tendency for higher inhibitory potential and lower substrate ability.
In this study, molecular docking was performed using the MOE application version 2015 [20] between the ligand in the form of curcumin and the 17 analogs of 2,5dibenzylidene cyclopentanone with CYP3A4, supporting the results of ADME screening for phase I metabolism parameters.CYP3A4 was employed as the receptor due to its role in catalyzing a significant portion of phase I metabolism in various drugs and being responsible for internal biotransformation, including hormones and various physiological glands.CYP3A4 is a P450 enzyme with the uniqueness of hydroxylating all positions of the aromatic carbons (ortho, meta, and para), thus rendering it widely catalytic.Through validation protocols, the visualization of interactions between the CYP3A4 receptor and the test ligands was obtained.By means of validated docking scores, predictive pIC50 values were obtained, indicating the inhibitory ability of CYP3A4 by curcumin and the 2,5-dibenzylidene cyclopentanone analogs (figure 8) (table 4).Potency of Inhibitory Concentration at 50% (pIC50) is a logarithmic measure used to represent the potency or inhibitory activity of a substance, typically a drug or chemical compound, in inhibiting a specific biological target, such as an enzyme or receptor [21].It quantifies the concentration of the substance required to achieve 50% inhibition of the target's activity.A lower pIC50 value indicates higher potency, meaning that a lower concentration of the substance is needed to achieve the desired inhibition.Out of the 18 test ligands, all compounds showed predictive pIC50 values.The highest predicted pIC50 values were observed for analogs compounds with codes B13, B14, B12, B15, followed by curcumin as the fifth compound with the highest pIC50 value.
Compounds with codes B13, B14, and B12 are analogs of 2,5-dibenzylidene cyclopentanone with alkyl hydrocarbon and hydroxyl substituents covering the ortho, meta, and para positions on the aromatic ring.The presence of these substituents, which exhibit the highest inhibition, is associated with the obstruction of CYP3A4 hydroxylation at those positions.On the other hand, the lowest pIC50 values were observed for compounds with codes B2 and B0, which only have hydrogen substituents and one hydroxyl group.This is associated with the availability of CYP3A4 hydroxylation space in the aromatic area of these compounds.The predictive pIC50 values obtained indicate that most of the 2,5dibenzylidene cyclopentanone analog compounds are still inclined to inhibit CYP3A4.This result correlates positively with the screening results using both ADME online servers, which were used to predict the probability of inhibition of the test compounds against the P450 isoform.Therefore, further studies are needed regarding the inhibition potential and synthesis design of other mono-ketone analogs of curcumin, taking into consideration their pharmacokinetic factors.
Based on the research conducted by Kaur et al. in 2016 on the interaction of ritonavir with CYP3A4 as an inhibitor through pharmacophore modeling, the presence of Phe (phenylalanine) residue in the hydrophobic interaction near the heme as a key amino acid residue was identified as a marker of the inhibitory interaction by the ligand to the CYP3A4 receptor.Upon validation with a known ligand, in ligand X6V (tert-butyl  9), a similar interaction with the CYP3A4 receptor 3UA1 was obtained [22].In curcumin (figure 10) the observed inhibiton interaction involves the aromatic groups of kurkumin interacting with HEM (P450 cofactor).In the compound B2 (figure 11), which lacks substituents, the observed inhibitory interaction involves the aromatic groups of B2 interacting with HEM (P450 cofactor).In the compound B1 (figure 12), commonly known as PGV-0 (Pentagamavunon-0) with the same substituents as kurkumin but in the form of a cyclic monocarbonyl, the observed inhibitory interaction involves HEM (P450 cofactor) interacting with the aromatic structure.In the compound B14, one of the compounds with the highest predictive pIC50 value, the interaction observed is similar to the known ligand X6V, involving the interaction between the Phe residue and the aromatic structure (figure 25).

Elimination Capability
The elimination parameters examined in this study are clearance (Cl) and half-life (T1/2).Both parameters are inversely related to the drug.Clearance is a pharmacological parameter that quantifies the irreversible release rate of a drug from the body [23].Since drugs can be eliminated from the body by various organs (kidneys, BIO Web of Conferences 75, 04002 (2023) https://doi.org/10.1051/bioconf/20237504002BioMIC 2023 liver, lungs, saliva, mammary glands, etc.), different types of clearance are distinguished based on these organs.Some of them include renal clearance (CLr) and hepatic clearance (CLh) [24].On the other hand, the half-life is the time required for the drug concentration (usually in blood or plasma) to decrease by half from its initial value [25].
Among the three ADME screening online servers, only ADMETLab 2.0 provides data on clearance (klirens) and half-life (t1/2) as elimination parameters.The lower the T1/2 value, the shorter the elimination time.Conversely, the lower the clearance value, the longer the elimination time from the body.Screening results with ADMETLab 2.0 (table 2) show lower T1/2 values compared to curcumin.However, the clearance values of some 2,5-Dibenzylidene cyclopentanone analogs are lower than curcumin, including compounds B3, B6, B9, B12, B13, B14, and B16.These low clearance values are associated with the presence of hydrocarbon chain substituents in compounds such as B6 (t-C4H9 substituent), B12 (two C2H5 substituents), B13 (two i-C3H7 substituents), and B14 (two i-C4H9 substituents).It is strongly suspected that the lower solubility of these nonpolar substituents in water makes it difficult for them to dissolve during glomerular filtration.Additionally, the presence of long-chain hydrocarbon substituents results in a slower metabolism, higher binding to plasma proteins, and stronger binding to fatty tissues, leading to longer elimination times (clearance).However, contrary to the nonpolar-hydrophobic logic, the presence of halide substituents, such as Cl in compounds B3 and B16, actually shows low clearance values.This is strongly suspected to be associated with a series of pharmacological processes in the body that render these elements less polar and unable to have high clearance abilities, such as protonation (hydrogen bonding), electrostatic interactions with oppositely charged groups like carboxylate (-COO), and formation of double bonds.These chemical processes are believed to originate from enzymatic mechanisms and various other physiological processes in the body.These physiological processes can involve ion exchange and transport, which involve various specific transporters and ion channels present in cell membranes in various tissues and organs, such as the kidneys, inteFstines, and nervous system [26] [27].One of the physiological processes strongly suspected is the firstphase metabolism by the P450 complex, which oxidizes drugs through hydroxylation of aromatic groups to increase the drug's polarity.The presence of halide as a substituent on the aromatic group in some analogs of 2,5dibenzylidene cyclopentanone leads to the inhibition of hydroxylation by P450 [28].This results in the inhibition of hydroxylation, causing the drug to become less polar and unable to be rapidly released from the hydrophilic fraction of the body.As a consequence, the clearance value of these compounds becomes low.

CONCLUSIONS
The results of pharmacokinetic character screening for 2,5-dibenzylidene cyclopentanone analogs indicate various predictive characteristics related to the physicochemical properties and ligand-receptor interaction abilities between the test compounds and proteins involved in drug pharmacokinetics in the body.Generally, the pharmacokinetic properties of 2,5dibenzylidene cyclopentanone analogs that show improvements in drug-likeness compared to curcumin are parameters commonly related to physicochemical properties such as human intestine absorption (HIA), BBB permeability, Volume of Distribution (VD), total clearance value (Cl), and half-life (T1/2).On the other hand, the pharmacokinetic properties that tend to decrease are ligand-receptor interactions, such as an increasing trend in P-Glycoprotein inhibition, strong binding with plasma proteins (Plasma Protein Binding/PPB), and a higher probability of predictive P450 inhibition compared to curcumin.The interactions with BPP and Pglycoprotein are qualitatively demonstrated through interactions with key amino acid residues via molecular docking simulations.The docking process conducted is a validated process, including ligand and protein preparation, pose and scoring validation, and docking process following the validated results. he docking results of the CYP3A4 receptor with curcumin and 2,5dibenzylidene cyclopentanone analogs indicate a predictive inhibitory ability, necessitating further in vitro or in vivo research to validate and provide a basis for the synthesis design of subsequent curcumin derivatives.
Although the in silico methods used in this study provide comprehensive pharmacokinetic properties and are reasonably correlated with the theoretical physicochemical properties of the drugs, these various parameters need to be validated in vitro and in vivo to prove their accuracy in the real physiological environment of the body.The in silico profiles developed in this study are expected to streamline future research, but further validation through experimental studies is still necessary.

Figure 2 :
Figure 2 : Main Frame of 2,5-Di2,5dibenzylidenecyclopentanone pharmacokinetic properties (Good pharmacokinetic characteristics are already present in curcumin and are further enhanced in the analog 2,5-dibenzylidene cyclopentanone) In drug development, an important determinant of the suitability of a drug molecule is the Rule of Five (RO5) proposed by Lipinski.The use of this rule is to predict the drug-likeness of various chemical compounds with a specific biological activity designed for orally administered drugs.Drug likeness: According to the RO5, good physicochemical properties of a drug are indicated by a molecular weight (MW) <500 g/mol, Log P value <5 (representing hydrophobicity), hydrogen bond donors (HBD) <5 sites, hydrogen bond acceptors (HBA) <10 sites, and polar surface area ≤140 Å.In further studies, an additional parameter was found: the presence of rotatable bonds (RB) <10.Deviations from more than one RO5 rule indicate the possibility of low gastrointestinal absorption[7] [8].

Figure 5 :
Figure 5 : The binding of Plasma Protein (PPB) Albumin with PDB 6WUW demonstrates interactions between albumin and the native ligand (green ribbon), curcumin (orange ribbons), and one of the analogs of 2,5-Dibenzylidene cyclopentanone (B14 shown in purple ribbon

Figure 7 .
Figure 7 .The hydroxylation positions of P450 isoforms on the aromatic ring

Figure 8 .
Figure 8.The docking results of the CYP3A4 receptor (3UA1) with curcumin (orange ribbon) and compound B14 (purple ribbon) in the best pose (visualized by MOE 2015)

Figure 10 .
Figure 10.The interaction observed between CYP3A4 (3UA1) and kurkumin involves the HEM (P450 cofactor) interacting with the aromatic groups of kurkumin

Figure 11 .
Figure 11.The interaction observed between CYP3A4 (3UA1) and B2 (analog without substituents) involves the HEM (P450 cofactor) interacting with the aromatic groups of B2

Table 4 .
Docking scores of curcumin and 2,5-dibenzylidene cyclopentanone analogs along with the conversion to pIC50 values based on the validated GBV/WSA DG scoring equation.