In Silico Identification of Promising PDE5 Inhibitors Against Hepatocellular Carcinoma Among Natural Derivatives: A Study Involving Docking and ADMET Analysis

Abstract

Hepatocellular carcinoma (HCC) represents a significant worldwide health challenge due to its high mortality rate, underscoring the need for advanced therapeutic strategies. This study employs a computer-based method to identify potential phosphodiesterase 5 (PDE5) inhibitors from a library of approved IBS_Scaff 532 natural compounds. PDE5 inhibitors have gained attention for their potential anti-tumor effects. Using molecular docking simulations, the researchers assessed how well these compounds bind to the PDE5 enzyme, which regulates cellular cGMP pathways. Additionally, ADMET profiling predicted the pharmacological and safety properties of candidate inhibitors. Notably, compounds like IBS_NC-0322 and IBS_NC-0320 exhibited favorable ADMET properties and strong binding affinities. These findings suggest their potential as therapeutic agents for treating HCC. While in silico methods serve as valuable screening tools, subsequent experimental validation and clinical trials are essential for confirmation.

Keywords

ADMET - cGMP - hepatocellular carcinoma (HCC) - molecular docking - phosphodiesterase 5

Introduction

Carcinoma accounts for 80 to 90% of all cancer diagnoses, making it the most common type of cancer. It develops in epithelial tissue, which forms protective linings on the skin, interior passageways (such as the esophagus), and most organs in the body. Cancers affecting the skin, breasts, prostate gland, kidneys, liver, pancreas, lungs, head, or neck fall under the category of carcinomas [1]. HCC is a malignancy with high morbidity and mortality, primarily affecting the liver. Risk factors for HCC include hepatitis B virus (HBV) and hepatitis C virus (HCV) infections, alcohol consumption, obesity, aflatoxin exposure, and type 2 diabetes [2]. In the initial stages, hepatocellular carcinoma (HCC) may be managed through options like surgical removal, organ transplant, or localized treatments. Yet, the diagnosis often occurs when the disease has progressed, resulting in a reduction of viable treatment methods [3].

Phosphodiesterase-5 (PDE5) inhibitors are a class of compounds that modulate cyclic nucleotide signaling pathways. They have gained attention in cancer research due to their potential to impact tumor growth and immune responses. Let’s explore the role of PDE5 inhibitors in HCC [4] [5]. Initially developed for managing systemic high blood pressure and chest pain, sildenafil is now widely marketed for the treatment of erectile dysfunction. It functions as an effective vasodilator, enhancing the action of nitric oxide and the signaling pathway of cyclic guanosine monophosphate (cGMP), similar to other PDE5 inhibitors such as tadalafil and vardenafil [6] [7]. Beyond their primary use, PDE5 inhibitors offer therapeutic potential for several diseases, including HCC [8]. Evidence suggests that PDE5 inhibitors may have anti-cancer effects in various malignancies. These agents have the potential to diminish the size of tumors, trigger apoptosis, and enhance the efficacy of existing chemotherapeutic drugs. Research has been conducted on their application in treating various cancers, including those of the breast, colorectal region (CRC), ovaries, and prostate [9].

The biological implications of PDE5 inhibitors in cancer remain partially understood. The molecular impact of these compounds is crucial in a variety of diseases such as renal disorders, diabetes mellitus, and various forms of cancer [10]. In particular, they have been investigated for their potential in glioblastoma multiforme (GBM) treatment, where they enhance brain tumor permeability and improve the effectiveness of standard chemotherapeutics. Although preclinical results are encouraging, there are currently limited clinical trials assessing the effectiveness of PDE5 inhibitors in treating glioblastoma (GBM). Additional research is necessary to determine their specific impact, taking into account the molecular diversity present in different patients [11].

In silico approaches, such as docking studies and ADMET analysis, a crucial role in drug discovery by combining computational predictions with experimental validation. They enhance efficiency, reduce costs, and contribute to finding viable drug candidates [12]. ADMET (Absorption, Distribution, Metabolism, Elimination, Toxicity) studies evaluate a drug candidate’s behavior in the body, including bioavailability, distribution, metabolism, excretion, and toxicity. Understanding ADME properties allows drug developers to make informed decisions about drug candidates, ensuring regulatory approval [13] [14].

Researchers continue to explore the therapeutic potential of PDE5 inhibitors in HCC. Combining PDE5 inhibitors with other targeted therapies or immunotherapies may enhance their efficacy. In summary, the intersection of PDE5 inhibitors, ADMET analysis, and innovative therapeutic approaches holds promise for advancing HCC treatment. As we unravel the complexities of HCC, these strategies pave the way toward better outcomes for patients.

Materials and Methods

Selection of Natural Compounds for Screening

Absolutely, the selection of a diverse library of natural compounds is a strategic step in your screening process. The IBS scaffolds library, with its 532 natural compounds, offers a broad spectrum of chemical structures and biological activities, which is indeed essential for identifying novel PDE5 inhibitors.

The diversity in the library increases the likelihood of finding compounds that can interact effectively with the PDE5 enzyme. Each compound’s unique chemical structure may offer different modes of binding to the enzyme, potentially leading to the discovery of a potent inhibitor with a novel mechanism of action.

Moreover, natural compounds often have the advantage of being structurally complex and biologically optimized through natural selection, which can translate into better pharmacological profiles and lower toxicity. This makes them excellent candidates for drug development programs, especially when looking for inhibitors of well-established targets like PDE5.

By utilizing in silico screening methods, I can efficiently evaluate these compounds against the PDE5 enzyme, represented by the crystal structure PDB ID 6L6E. This approach not only saves time and resources but also allows for a more focused investigation of the most promising candidates based on their docking scores and predicted ADMET profiles.

The study begins by carefully selecting IBS scaffolds Natural compounds (532) library (https://www.ibscreen.com/scaffolds-download) that will undergo in silico screening [15]. These compounds are likely to have diverse chemical structures and biological activities. The goal is to identify potential PDE5 inhibitors among these natural derivatives [16].

In summary, we choice to screen a diverse set of natural compounds from the IBS scaffolds library is a sound scientific strategy that aligns well with the goals of your research in identifying new PDE5 inhibitors.

Protein-protein interactions (PPI)

Indeed, databases like STRING are invaluable resources in the field of drug development, particularly when it comes to understanding protein-protein interactions (PPIs). These interactions are crucial for the cellular functions and are often the targets for therapeutic intervention. Here’s how databases like STRING contribute to the process of Comprehensive Data, Functional Characterization, Interaction Prediction, Network Analysis, Drug Discovery and Biases and Enrichment. STRING integrates known and predicted PPI data from multiple sources, providing a vast and comprehensive dataset that covers a wide range of organisms. It allows for the functional characterization of proteins and the identification of potential drug targets by highlighting functional associations between proteins. STRING uses various computational methods to predict interactions, which can be particularly useful for identifying novel interactions that have not been experimentally observed yet. The database enables researchers to visualize and analyze protein networks, which can reveal new insights into how proteins interact within a biological system. By understanding the interaction networks, researchers can identify critical nodes and interactions that can be targeted by drugs, aiding in drug target discovery and repurposing efforts. STRING also provides tools for detecting potential biases in the data and for performing enrichment analysis, which can help in prioritizing targets for further investigation [17] [18].

PPIs are essential for comprehending protein functions and acquiring knowledge of biochemical and/or metabolic processes. The STRING database collects, evaluates, and combines all publicly available datasets of protein-protein interaction data, augmenting them with computational forecast [19] [20].

The use of STRING and similar databases is a key step in your methodology for identifying PDE5 inhibitors. By analyzing the PPIs, you can gain a deeper understanding of the PDE5 enzyme’s role and interactions within the cell, which in turn can inform the selection of compounds that may modulate its activity. This integrative approach, combining computational predictions with experimental validation, is at the forefront of modern drug discovery and development.

PDE5 inhibitors against Hepatocellular Carcinoma Activity Prediction

The PDE5 profile (Vasodilator/peripheral and HCC) of the top four IBS natural compounds as PDE5 inhibitors against HCC was predicted using the PASS online tool (http://www.way2drug.com/passonline/). Compounds exhibiting a probable activity (Pa) value exceeding 0.3 underwent molecular docking analysis. A Pa value exceeding 0.3 suggests that the chemical possesses promise as an in silico PDE5 inhibitor for HCC [21].

In Silico Methods for Identifying PDE5 Inhibitors

In silico methods involve computational techniques to predict interactions between IBS scaffolds Natural compounds (532) and biological targets (such as PDB ID 6L6E). This in silico approach for identifying potential PDE5 inhibitors is well-founded and employs a variety of computational techniques to ensure a thorough evaluation of the natural compounds. Here’s a detailed look at each step:

• Molecular Docking

  Utilizing the Schrodinger suite (Schrodinger Release: Maestro 13.9, Schrodinger, LLC, New York, NY) for docking simulations is indeed a robust choice. The crystal structure of human PDE5 (PDB ID 6L6E) is a critical target for docking The Natural compounds. The IBS scaffolds Natural compounds (532) library are virtually docked into the active site of the PDE5 enzyme cGMP like PDBID 6L6E. Obtain the crystal structure of human PDE5 (PDBID 6L6E) from the protein databank (https://www.rcsb.org/) [22] [23]. Prepare the protein wizard in the Glide program (Schrödinger: Protein Preparation Wizard; Epik, Schrödinger, LLC, New York, NY) by removing water molecules, adding missing atoms, and optimizing its geometry.

• Scoring and Ranking

  The docking scores are essential as they provide a quantitative measure of each compound’s binding affinity to PDE5. Compounds with higher scores are indicative of stronger interactions, which can be promising candidates for further analysis [24].

  By examining the binding modes and interactions, such as hydrogen bonds and hydrophobic interactions, you can gain valuable insights into the inhibitory mechanisms of the natural compounds. This analysis is crucial for understanding how these compounds might inhibit PDE5 [25].

• Molecular docking simulation by PLANTS

  PLANTS stands for Protein-Ligand ANT System. This docking tool is aimed at virtual screening and predicting binding modes for ligands on target proteins. It relies on algorithms of ant colony optimization with the aim of analyzing different conformation possibilities of ligands in order to identify the most energy-favored poses of ligands. The approach is very well adapted for large virtual screens of compound libraries, with high precision in ligand docking [26].

  After the Protein-Ligand prediction, calculate the RMSD and Iterations using PLANTS. The docking score gives an optimal docked conformation. Look through your molecular dynamics trajectory file to get the RMSD. Iterations in computational simulations (like molecular dynamics) refer to repeating simulation steps to model molecular interactions over time. The value one attains here measures by how much the ligand-protein pair has moved from their docked starting position.

• Virtual ADMET (Absorption, Distribution, Metabolism, Elimination, Toxicity) Prediction

  Tools like pkCSM and SwissADME are instrumental in predicting the pharmacokinetic properties and safety profiles of the top docked compounds. These predictions are vital for assessing the potential of these compounds as safe and effective drugs. Using pkCSM (https://biosig.lab.uq.edu.au/deeppk/) and SwissADME ( http://www.swissadme.ch/index.php#), we determined the ADMET characteristics of the top docked compounds [27] [28].

• Toxicity analysis

The determination of toxicity is part of drug discovery and is an important step for predicting the possible adverse effects of drugs and chemicals on biological organisms. Toxicity research for the selected IBS_Natural compounds can be carried out by ProTox 3.0 (https://tox.charite.de/protox3/), which is most commonly used in silico to predict various toxicological end points. ProTox 3.0 is a proficient instrument for forecasting the toxicity of the chosen IBS_Natural substances. By offering predictions on acute toxicity (LD50), hepatotoxicity, carcinogenicity, and other toxicological parameters, it facilitates a more accurate assessment of the potential detrimental consequences of these chemicals. This facilitates subsequent biological assessments and the secure advancement of treatment medicines for HCC.

General Approach

Toxicity analysis is evaluating the probability that a substance may adversely affect a biological organism, either in an acute or chronic manner. Toxicity outcomes generally encompass:

• LD50 (Lethal Dose 50%): The amount of dose that kills 50% of a test population. This is another indicator of acute toxicity. It gives an estimated value of the dose in mg/kg necessary to kill 50% of a test population and falls within the toxicity classes of 1 (most toxic) to 6 (least toxic). Toxicity Class: ProTox 3.0 classify the chemical depending on the LD50 in a specific toxicity class:

  • Class 1: Death if swallowed (LD50≤5 mg/kg)

  • Class 2: Death if swallowed (5<LD50≤50 mg/kg)

  • Class 3: Toxic if swallowed (50<LD50≤300 mg/kg)

  • Class 4: Harmful if swallowed (300<LD50≤2000 mg/kg)

  • Class 5: May be harmful (2000<LD50≤5000 mg/kg)

  • Class 6: Non-toxic (LD50>5000 mg/kg)

  Other hazard estimates for hepatotoxicity, carcinogenicity, mutagenicity and immunotoxicity are also reported as yes/no where applicable, with associated scores.

• Hepatotoxicity: The ability of the compound to cause damage to the liver.

• Carcinogenicity: Cancer causing ability of the chemical.

• Mutagenicity: The chemical's ability to cause mutations in the DNA.

• Immunotoxicity: The chemical's ability to damage the immune system.

• Cytotoxicity: Ability of the chemical to cause cell damage.

• Comparison with Known Inhibitors

Comparing the identified compounds with known PDE5 inhibitors, such as sildenafil, is a smart strategy. This step helps validate the potential efficacy of the new inhibitors by ensuring they share similar mechanisms of action with established drugs [29] [30]. The identified compounds are compared with known PDE5 inhibitors (like sildenafil) to validate their potential [31].

Recent advancements in molecular docking have highlighted the importance of considering protein flexibility, ligand sampling, and scoring functions, which are key aspects in the accuracy of docking simulations [32] [33] [34]. Additionally, the comparison of ADMET prediction tools like pkCSM and SwissADME has shown that both tools perform well, with several trials indicating their effectiveness in providing accurate and comprehensive predictions [28].

Known PDE5 inhibitors, such as sildenafil (Viagra), tadalafil (Cialis), and vardenafil (Levitra), work by blocking the PDE5 enzyme to prevent it from functioning, which relaxes blood vessels and increases blood flow. This mechanism is particularly effective in treating conditions like erectile dysfunction and pulmonary hypertension [31].

This in-silico methodology combines computational predictions with the potential for experimental validation, which is a hallmark of modern drug discovery. It’s a strategic approach that maximizes the chances of identifying novel PDE5 inhibitors from natural compounds.

Criteria for Selecting Promising PDE5 Inhibitors

The selection criteria for potential PDE5 inhibitors are critical in the journey of drug discovery. Docking scores serve as a quantitative indicator of how well a ligand fits and binds to its target protein. These scores are key in identifying compounds with a higher likelihood of effective binding to the PDE5 enzyme. Elevated docking scores are indicative of more robust interactions, which may lead to enhanced inhibitory effects [35] [36].

ADMET, an acronym for Absorption, Distribution, Metabolism, Excretion, and Toxicity, outlines the properties vital for evaluating a drug’s safety and effectiveness. Predicting these characteristics early on can greatly diminish the chances of a drug failing in later development stages by spotlighting compounds with optimal pharmacokinetic and safety profiles [37] [38].

Evaluating the structural similarities between new compounds and established PDE5 inhibitors, like sildenafil, offers valuable insights into their possible effectiveness. Compounds that structurally mimic successful drugs are likely to exhibit comparable actions, providing a solid foundation for the discovery of new medicinal agents [39].

By applying these selection criteria, researchers can efficiently filter through a large array of potential inhibitors to identify a select group of promising candidates. This methodical approach not only facilitates the discovery process but is also in line with the objective of uncovering safe and efficacious PDE5 inhibitors for medical applications.

The methodology combines computational predictions with experimental validation, allowing researchers to efficiently identify potential drug candidates. By leveraging natural derivatives and in silico techniques, the study aims to contribute to the field of HCC treatment.

Result

The functional interaction analysis network of proteins

The STRING tool was used to conduct co-expression analysis, which provided co-expression scores based on protein coregulation and the pattern of RNA expression documented by Proteome HD shown in ([Fig. 1]).

Fig. 1 Co-expression and the protein-protein network analysis. STRING 11.5 was used to conduct a protein-protein network and co-expression analysis of PDE5 (cGMP). The highlighted edges show several types of evidence used to forecast the relationships.

There aren't many direct PDE5 inhibitors, despite the wealth of research showing PDE5 (cGMP) functions in immunity and inflammation via interactions with other proteins. Therefore, this study aimed to identify natural derivatives as PDE5 inhibitors.

PDE5 inhibitors against Hepatocellular Carcinoma Activity Prediction

The PASS Online program was applied to predict the Pa values of PDE5 inhibitors, considering the vasodilator, hepatoprotective, and chemopreventive activities of the selected IBS natural compounds against HCC, as reported in [Table 1]. IBS_NC-0322 showed limited activity predictions for hepatoprotective (Pa=0.375) and chemopreventive (Pa=0.411) activities. IBS_NC-0320 strongly exerted hepatoprotective activities (Pa=0.421) and had limited vasodilatory activities. IBS_NC-0505b had poorer predictive ability in all categories. IBS_NC-0502 showed the highest degree of chemopreventive activity (Pa=0.476) with also high potential of hepatic protection.

Table 1 Pa prediction activity of unknown PDE5 IBS_Natural Compounds against HCC.

IBS_Natural Compounds ID	Pa value of Vasodilator activity like tadalafil	Pa value of Hepatoprotectant activity	Pa value of Chemopreventive activity
IBS_NC-0322	0.304	0.375	0.411
IBS_NC-0320	0.328	0.421	0.361
IBS_NC-0505b	0.328	0.335	0.175
IBS_NC-0502	0.375	0.398	0.476

Compounds having Pa values>0.3 were further subjected to molecular docking study. These results implied them to be PDE5 inhibitors for the treatment of hepatocellular carcinoma.

Identification of top potential PDE5 Protein

The identification of the top potential PDE5 Protein is a critical step in drug discovery for diseases where the regulation of cyclic guanosine monophosphate (cGMP) is beneficial, such as erectile dysfunction and certain cardiovascular conditions. Recent studies have also explored the role of PDE5 inhibitors in cancer therapy, including Hepatocellular Carcinoma (HCC) [40].

Initially, we acquired the structure of PDE5 (PDBID: 6L6E) from the Protein Data Bank and visualized it using the Maestro module of the Schrödinger suite presented in ([Fig. 2a]). The protein preparation wizard then processed the acquired protein, clipping all chains except A shown in ([Fig. 2b]).

Fig. 2 Human PDE5 Protein. (a) Structure of PDE5 (PDBID: 6L6E) accurized from Protein Data Bank (b) Processed PDE5 structure with chain A by removing water and hetatm.

These findings highlight the importance of computational methods in the early stages of drug discovery, allowing for the efficient screening of large compound libraries to identify new therapeutic agents with potential applications in various diseases, including HCC.

Analysis of binding affinities and interactions

We used the Schrodinger suite's SiteMap module to identify the active site of the preprocessed proteins. We used SiteMap to find one possible active site for screening shown in ([Fig. 3]).

Fig. 3 Binding Sites acquired via SiteMap.

We found docking scores against PDE5 (PDBID: 6L6E) after effective virtual screening. The ligands' docking scores were utilized to determine their binding affinities with PDE5 enzyme [41]. 532 IBS scaffolds Natural compounds had docking scores of best − 12.6 to − 11.5. However, the top four molecules were selected for further research presented in ([Table 2]).

Table 2 Docking score of best 5 Recently docked IBS scaffolds Natural compounds via glide.

IBS_Natural Compounds ID and Structure	Source	Docking Affinity Score
	Triticum urartu	− 12.5789
	Oats of Triticum urartu	− 11.991
	Pancratium maritimum (Amaryllidaceae Alkaloids)	− 11.5611
	Avena sativa	− 11.5111

IBS_NC-0322, IBS_NC-0320, IBS_NC-0505b and IBS_NC-0502 is the top four unknown PDE5 inhibitors (IBS_Natural compounds). Their docking scores ranged from − 12.6 to − 11.5 tabulated in ([Table 1]).

Furthermore, docking studies showed that covalent energy and hydrogen bonds are very important for the best binding affinity of the chosen natural derivatives. The two- and three-dimensional docking poses showed that ligand binding with PDE5 enzyme cGMP is significantly influenced shown in ([Fig. 4] [5]).

Fig. 4 3D docking of best docked IBS scaffolds Natural compounds with PDE5 (cGMP). (a–e) The picture demonstrates the 3D pose of best docked unknown PDE5 inhibitors (IBS_Natural compounds) i. e., (a) IBS_NC-0322 (b) IBS_NC-0320 (c) IBS_NC-0505b (d) IBS_NC-0502 with PDE5 (cGMP) respectively.

Fig. 5 Ligand Interaction 2D docking of best docked IBS Scaff Natural Compounds with PDE5 (cGMP). (a–e) The picture demonstrates the 2D pose of best docked unknown PDE5 inhibitors (IBS_Natural compounds) i. e., (a) IBS_NC-0322 (b) IBS_NC-0320 (c) IBS_NC-0505b (d) IBS_NC-0502 with PDE5 (cGMP) respectively.

Comparison with known inhibitors

Comparing potential PDE5 inhibitors with known inhibitors is a crucial step in drug discovery. It helps establish the efficacy and safety profile of new compounds relative to existing treatments. Here’s a brief overview of how this comparison can be conducted.

• Binding Affinity

  Compare the binding affinities of the new inhibitors with those of known inhibitors like sildenafil, tadalafil, and vardenafil tabulated in ([Table 3]).

• Molecular Docking for unknown PDE5 inhibitors to known PDE5 inhibitors

Table 3 Comparison of Binding Affinity of best 5 docked unknown PDE5 inhibitors (IBS_Natural compounds) versus known PDE5 inhibitors.

S. No	Recently FDA Approved Compounds	Binding Affinity	Known PDE5 Inhibitors	Binding Affinity
1	IBS_NC-0322	− 12.5789	Tadalafil	− 10.5
2	IBS_NC-0320	− 11.991	Vardenafil	− 9.9
3	IBS_NC-0505b	− 11.5611	Sildenafil	− 9.8
4	IBS_NC-0502	− 11.5111	Avanafil	− 9.3

Perform molecular docking studies to visualize how the IBS_Natural Compounds new inhibitors fit into the PDE5 (PDBID: 6L6E) active site compared to known inhibitors like tadalafil. This can provide insights into the potential efficacy and selectivity of the new compounds, seen in the [Fig. 6].

Fig. 6 Comparison of molecular docking of high affinity recently docked PDE5 (IBS_Natural compounds) versus known PDE5 inhibitors.

By conducting a thorough comparison, we can determine the potential of new PDE5 inhibitors (IBS_Natural compounds) to advance into preclinical studies and eventually become part of the therapeutic options for conditions like HCC.

Molecular docking simulation or the ΔG scores from the PLANTS

[Fig. 7] Results from the MD simulations in PLANTS of ΔG, iterations, and time unit for each compound. IBS_NC-0322 has a ΔG value of − 73.31 after 259 iterations of time units 1.75. IBS_NC-0320 reached a ΔG score of − 76.6 after 293 iterations with 2.48-time units. IBS_NC-0505b has the lowest ΔG at -68.77 at only 125 iterations and within the minimum time units of 0.26. IBS_NC-0502 had the highest ΔG at -79.92 with 300 iterations and took a time of 2.55 units. The results would then indicate clearly that IBS_NC-0502 is indeed the most stable interaction and, therefore, holds more promise in further research work.

Fig. 7 The PLANTS MD simulation results: the best ΔG scores with iterations of the top four docked PDE5 inhibitors, such as IBS_NC_0322, IBS_NC_0320, IBS_NC_0505b, and IBS_NC_0502.

[Fig. 8] represents the RMSD values, which are indicative of the stability of the protein-ligand complexes during the whole course of molecular dynamics simulations. Among the four top-ranked compounds, IBS_NC-0320 has the least RMSD value that is 4.46212 Å, meaning its binding poses were the most stable, then came IBS_NC-0502 that had 5.29322 Å. On the other side, IBS_NC-0505b had RMSD value as 6.13962 Å and IBS_NC-0322 as 7.35703 Å that depicts higher fluctuation at the structural level. If time is taken as an indicator of the docking simulation of each compound, it will be clearly noticed from the following that IBS_NC-0505b corresponds to the shortest period. The results indicated that IBS_NC-0320 and IBS_NC-0502 are the most stable complexes.

Fig. 8 The PLANTS MD simulation results: RMSD with time of the top four docked PDE5 inhibitors, such as IBS_NC_0322, IBS_NC_0320, IBS_NC_0505b, and IBS_NC_0502.

ADMET (Absorption, Distribution, Metabolism, Elimination, Toxicity) studies

Drug discovery heavily relies on ADMET (Absorption, Distribution, Metabolism, Elimination, Toxicity) studies. These investigations help identify drug-like properties. Historically, up to 50% of drug candidates failed due to insufficient efficacy, while 40% failed due to toxicities. Notably, drugs like phenylpropanolamine hydrochloride and mibefradil were withdrawn from the market due to toxicity or drug interactions. Recognizing the impact of ADMET on overall drug quality and success rates, authorities and pharmaceutical companies now prioritize these studies early in drug development. In silico ADMET prediction is recommended for cost-effective screening. High-quality models allow simultaneous analysis and optimization of chemical effectiveness and druggability. The publicly accessible web server pkCSM and SwissADMET aids in ADMET profiling. Molecular weight significantly affects oral bioavailability, and all compounds fall within the suitable range for oral medications ²⁴, with IBS_NC-0505b being the lightest and IBS_NC-0502 the heaviest. A lower number of rotatable bonds can indicate a more rigid structure, which might affect the compound’s ability to fit into the active site of enzymes or receptors. IBS_NC-0320 and IBS_NC-0502 has more rotatable bonds, suggesting higher flexibility. Negative values indicate poor water solubility, which can impact absorption and distribution. IBS_NC-0505b appears to be the most water-soluble among the listed compounds. High percentages suggest good oral bioavailability. IBS_NC-0322, IBS_NC-0320, IBS_NC-0505b and IBS_NC-0502 show very high intestinal absorption. Four IBS compounds show Lower negative values suggest better skin permeability. Inhibition of P-glycoprotein can affect drug efflux and, thus, bioavailability. Four IBS compounds are inhibitors of both P-glycoprotein I and II, potentially affecting their distribution and elimination. Negative values suggest poor permeability. All compounds have negative values, with IBS_NC-0505b having BBB permeability, indicating it is less likely to cross into the CNS. None of the compounds acted as inhibitors of key cytochrome P450 enzymes. Skin permeation properties were favorable for all compounds. However, hepatotoxicity was observed for IBS_NC-0505b and IBS_NC-0502. Overall, these two IBS_NC-0322 and IBS_NC-0320 findings provide insights into the safety and potential therapeutic use of these IBS natural derivatives. Also, the chosen IBS_Natural compounds are within acceptable ranges regarding Drug properties like molecular weight, LogP, and bioavailability and thus are good candidates for further studies in HCC. IBS_NC-0502 is over the threshold of molecular weight for leadlikeness and, most probably, will require optimization. Synthetic accessibility scores show that all compounds are synthesizable, though IBS_NC-0502 is more probable to be prepared. Such evaluation of drug-likeness is a basis for further biological evaluation and for carrying out in vivo studies for the considered IBS_Natural compounds. according to the pkCSM and SwissADMET analysis of the molecules tabulated in ([Table 4] [5]).

Table 4 Drug-like properties of the best-selected PDE5 (IBS_Natural compounds).

Molecule properties:
Descriptor	IBS_NC-0322	IBS_NC-0320	IBS_NC-0505b	IBS_NC-0502
Molecular Weight	310.241	324.268	266.276	357.318
LogP	− 3.11822	− 2.72812	0.55058	− 0.48852
Rotatable Bonds	3	4	0	4
Acceptors	6	6	3	7
Donors	1	1	0	1
Surface Area	129.663	136.028	115.3	148.195
Leadlikeness	Yes	Yes	Yes	No; 1 violation: MW>350
Synthetic accessibility	3.98	4.06	4.27	3.42
Bioavailability score	0.56	0.56	0.55	0.56

Note: The SwissADME and RDKit portal scored drug-like parameters for ADMET, such as leadlikeness (implemented from Teague SJ. 1999 Angew. Chem. Int. ED. 250≤MW≥350), synthetic accessibility score typically ranges from 1 (easy to synthesize) to 10 (very difficult to synthesize), and bioavailability score (typical values being 0.55 for highly bioavailable compounds and 0.17 for poorly bioavailable compounds).

Table 5 ADMET-related drug-like parameters of the best-selected PDE5 (IBS_Natural compounds) using Deep-PK and SwissADME.

Pharmacokinetic Properties
Pharmacokinetics	IBS_NC-0322	IBS_NC-0320	IBS_NC-0505b	IBS_NC-0502
GI _Absorption	High	High	High	High
BBB _ Permeability	No	No	Yes	No
P-gp _substrate	No	No	Yes	No
CYP1A2 _inhibitor	No	No	No	No
CYP2C19 _inhibitor	No	No	No	No
CYP2C9 _inhibitor	No	No	No	Yes
CYP2D6 _inhibitor	No	No	No	No
CYP3A4 _inhibitor	No	No	No	No
Log Kp _ (skin permeation)	− 9.08 cm/s	− 8.95 cm/s	− 7.82 cm/s	− 7.19 cm/s
Total Clearance	0.55	0.631	0.916	0.594
Renal OCT2 _substrate	No	No	No	No
AMES _toxicity	No	Yes	No	No
Max. tolerated dose (human)	− 0.064	− 0.24	− 0.034	− 0.037
hERG I _inhibitor	No	No	No	No
hERG II _inhibitor	No	No	No	No
Oral Rat Acute _Toxicity (LD50)	2.021	1.999	1.992	2.424
Oral Rat Chronic _Toxicity (LOAEL)	1.164	1.166	1.987	2.112
Hepatotoxicity	No	No	Yes	Yes
Skin Sensitisation	No	No	No	No
T.Pyriformis toxicity	0.42	0.483	0.548	0.455
Minnow toxicity	3.232	2.957	2.337	2.264

Oral Toxicity analysis prediction

[Table 5] is a detailed prediction of the oral toxicity of four chosen PDE5 inhibitors (IBS_Natural compounds: IBS_NC-0322, IBS_NC-0320, IBS_NC-0505b and IBS_NC-0502 respectively obtained from the output of ProTox 3.0. Critical toxicological endpoints such as LD50, target organ toxicity, and general toxicity classifications were evaluated so that let in light of the safe profile of the drugs and their potential risks associated with oral exposure.

PDE5 (IBS_Natural products) manifests tendencies toward oral toxicity with a rather diversified safety profile. Compounds like IBS_NC-0505b show diminished acute toxicity risk whereas others like IBS_NC-0320 have higher potential risks of increased toxicity and carcinogenicity/nephrotoxicity. These findings clearly and urgently suggest a balance between therapeutic effectiveness and drug safety. Later work should aim to develop these drugs to reduce the toxicity associated with them, especially the carcinogenic and nephrotoxic potential of certain of these drugs, without losing their useful therapeutic properties.

Discussion

The in-silico investigations take into account the efficiency of natural PDE5 inhibitors against hepatocellular carcinoma, which is validated by various analytical techniques. Analytical tool used was STRING to look at co-expression and protein-protein interactions. It was very intriguing to discover patterns in PDE5, especially its association with cGMP signaling pathways, as shown in [Fig. 1]. A significant amount of research has been established between PDE5 and immunity/inflammation, but few compounds have direct inhibitors. Thus, finding natural leads that can bind this enzyme is crucial. According to the PASS Online program, IBS_NC-0320 and IBS_NC-0502 were good candidates. They were supposed to have hepatoprotective and chemopreventive activities, that further was accredited by molecular docking scores. Docking experiments further showed that these natural chemicals possess strong binding affinities that is supported by the 2D and 3D images of protein-ligand interactions as shown in [Fig. 4] and [5]. Moreover, comparison of these compounds along with known inhibitors like tadalafil and sildenafil showed that they have potential therapeutic value since they show more binding affinities as shown in the [Table 3]. The PLANTS MD simulation was able to identify the optimal docking positions and interaction with PDE5 for the four most significant IBS natural compounds: IBS_NC-0322, IBS_NC-0320, IBS_NC-0505b, and IBS_NC-0502. The strongest binding affinity and lowest ΔG score (-79.92 kcal/mol) was indicated for the IBS_NC-0502, indicating that it had strong interactions. IBS_NC-0320 is at second position. Furthermore, the lowest RMSD value is 4.46 as shown in [Fig. 7] [8]. It showed that the IBS NC 0320 complex had the most stable form. ADMET and oral toxicity tests were conducted that showed that these compounds are drug-like candidates; however, IBS_NC-0505b and IBS_NC-0502 need optimization with respect to hepatotoxicity ([Table 4] [5] [6]). The results point out that IBS_NC-0502 and IBS_NC-0320 show promise as leads for the investigation as potential PDE5 inhibitors against Hepatocellular Carcinoma.

Table 6 Oral toxicity prediction of the best-selected PDE5 (IBS_Natural compounds) using ProTox 3.0.

Oral toxicity Properties
Organ Target	IBS_NC-0322	IBS_NC-0320	IBS_NC-0505b	IBS_NC-0502
LD50	470 mg/kg	105 mg/kg	2000mg/kg	1200 mg/kg
Toxicity Class	4	3	4	4
Hepatotoxicity	Inactive	Inactive	Inactive	Inactive
Neurotoxicity	Inactive	Inactive	Active	Inactive
Nephrotoxicity	Active	Active	Inactive	Active
Respiratory toxicity	Active	Active	Active	Active
Cardiotoxicity	Inactive	Inactive	Inactive	Inactive
Carcinogenicity	Active	Active	Active	Inactive
Immunotoxicity	Inactive	Inactive	Inactive	Active
Mutagenicity	Inactive	Inactive	Inactive	Inactive
Cytotoxicity	Inactive	Inactive	Inactive	Inactive
BBB-barrier	Active	Active	Active	Active
Ecotoxicity	Active	Active	Active	Inactive
Clinical toxicity	Inactive	Inactive	Inactive	Inactive
Nutritional toxicity	Active	Active	Active	Inactive

Conclusion

This study on the identification of new PDE5 inhibitors for the treatment of Hepatocellular Carcinoma (HCC) using in silico methods has yielded promising results. The in-silico screening identified several potential PDE5 inhibitors with cGMP that could be repurposed for HCC treatment. We have discovered several promising naturally occurring PDE5 inhibitors that can combat hepatocellular carcinoma. We have applied multiple computational techniques, including co-expression network analysis, activity prediction, molecular docking, and molecular dynamics simulations. Of the four compounds investigated in this study, IBS_NC-0502 and IBS_NC-0320 showed the greatest promise, with high anticancer activities and hepatoprotective effects, strong affinities for ligand-protein binding, and persistent interactions between proteins and ligands. The only toxicity concerns are related to hepatotoxicity and carcinogenic potential, as estimated from ADMET and oral toxicity. However, despite these issues, these drugs still present considerable promise. Further optimization and studies in vivo will be necessary to enhance the therapeutic benefits of these compounds, reduce their toxicities, and establish them as effective treatments for HCC.

Conflict of Interest

No potential conflict of interest was reported by the authors.

Acknowledgment

The present study was supported by a grant from the Research Director of Patel college of Pharmacy, MPU, Bhopal, India and MET Faculty of Pharmacy, MIT, Moradabad, UP. India.

• References

• 1 Kumar A, Rajput DS, Gupta N. Phosphodiesterase (PDE) as a novel target to suppress Carcinoma: a future perspective. YMER Dig 2024; 23: 833-854
• 2 Cruz-Burgos M, Losada-Garcia A, Cruz-Hernández CD. et al. New Approaches in Oncology for Repositioning Drugs: The Case of PDE5 Inhibitor Sildenafil. Front Oncol 2021; 11: 627229
• 3 Tabori NE, Sivananthan G. Seminars in IR Liver Oncology: Treatment Options for Early-Stage Hepatocellular Carcinoma. Semin Inter Rad 2020; 37: 448
• 4 Barone I, Giordano C, Bonofiglio D. et al. Phosphodiesterase type 5 and cancers: progress and challenges. Oncotarg 2017; 8: 99179
• 5 ElHady AK, El-Gamil DS, Abdel-Halim M. et al. Advancements in Phosphodiesterase 5 Inhibitors: Unveiling Present and Future Perspectives. Pharmaceu 2023; 16: 1266
• 6 Terrett NK, Bell AS, Brown D. et al. Sildenafil (Viagra (TM)), a potent and selective inhibitor of type 5 CGMP phosphodiesterase with utility for the treatment of male erectile dysfunction. Bioorg Med Chem Lett 1996; 6: 1819-1824
• 7 Ghofrani HA, Osterloh IH, Grimminger F. Sildenafil: from angina to erectile dysfunction to pulmonary hypertension and beyond. Nat Rev Drug Dis 2006; 5: 689
• 8 Corbin JD, Francis SH. Cyclic GMP phosphodiesterase-5: Target of sildenafil. J Bio Chem 1999; 274: 13729-13732
• 9 Huang W, Sundquist J, Sundquist K. et al. Phosphodiesterase-5 inhibitors use and risk for mortality and metastases among male patients with colorectal cancer. Nat Comm 2020; 11: 3191
• 10 Xu L, Sun L, Su P. et al. Identification of Potential Inhibitors of PDE5 based on Structure-based Virtual Screening Approaches. Curr Comp Aid Drug Des 2023; 19: 234-242
• 11 Yalamarty SSK, Filipczak N, Li X. et al. Mechanisms of Resistance and Current Treatment Options for Glioblastoma Multiforme (GBM). Can (Bas) 2023; 15: 2116
• 12 Agamah FE, Mazandu GK, Hassan R. et al. Computational/in silico methods in drug target and lead prediction. Brief Bio 2020; 21: 1663
• 13 Guan L, Yang H, Cai Y. et al. ADMET-score – a comprehensive scoring function for evaluation of chemical drug-likeness. MedChemComm 2019; 10: 148
• 14 Wan H. What ADME tests should be conducted for preclinical studies?. ADMET DMPK 2013; 1: 19-28
• 15 Patil VR, Dhote AM, Patil R. et al. Identification of structural scaffold from interbioscreen (IBS) database to inhibit 3CLpro, PLpro, and RdRp of SARS-CoV-2 using molecular docking and dynamic simulation studies. J Biomol Struct Dyn 2023; 41: 13168-13179
• 16 Franceschin M, Cianni L, Pitorri M. et al. Natural Aromatic Compounds as Scaffolds to Develop Selective G-Quadruplex Ligands: From Previously Reported Berberine Derivatives to New Palmatine Analogues. Mol 2018; 23: 1423
• 17 Szklarczyk D, Kirsch R, Koutrouli M. et al. The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res 2023; 51: D638-D646
• 18 Szklarczyk D, Gable AL, Nastou KC. et al. The STRING database in 2021: customizable protein–protein networks, and functional characterization of user-uploaded gene/measurement sets. Nucleic Acids Res 2021; 49: D605-D612
• 19 Basar MA, Hosen MF, Kumar Paul B. et al. Identification of drug and protein-protein interaction network among stress and depression: A bioinformatics approach. Inform Med Unlocked 2023; 37: 101174
• 20 Ding Z, Kihara D. Computational identification of protein-protein interactions in model plant proteomes. Sci Rep 2019; 9: 8740
• 21 Tjitda PJP, Nıtbanı FO, Parıkesıt AA. et al. In Silico Investigation of Tropical Natural Product for Wild-Type and Quadrupole Mutant PfDHFR Inhibitors as Antimalarial Candidates. Trop J Nat Prod Res 2024; 8: 6208-6217
• 22 Hsieh CM, Chen CY, Chern JW. et al. Structure of Human Phosphodiesterase 5A1 Complexed with Avanafil Reveals Molecular Basis of Isoform Selectivity and Guidelines for Targeting α-Helix Backbone Oxygen by Halogen Bonding. J Med Chem 2020; 63: 8485-8494
• 23 Killari KN, Polimati H, Prasanth DSNBK. et al. Salazinic acid attenuates male sexual dysfunction and testicular oxidative damage in streptozotocin-induced diabetic albino rats. RSC Adv 2023; 13: 12991-13005
• 24 Chen CY, Chang YH, Bau DT. et al. Discovery of potent inhibitors for phosphodiesterase 5 by virtual screening and pharmacophore analysis. Acta Pharmacol Sin 2009; 30: 1186
• 25 Varma AK, Patil R, Das S. et al. Optimized Hydrophobic Interactions and Hydrogen Bonding at the Target-Ligand Interface Leads the Pathways of Drug-Designing. PLoS One 2010; 5: e12029
• 26 Korb O, Stützle T, Exner TE. PLANTS: Application of Ant Colony Optimization to Structure-Based Drug Design. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2006; 4150: 247-258
• 27 Pires DEV, Blundell TL, Ascher DB. pkCSM: Predicting Small-Molecule Pharmacokinetic and Toxicity Properties Using Graph-Based Signatures. J Med Chem 2015; 58: 4066-4072
• 28 Azzam KAL. SwissADME and pkCSM Webservers Predictors: an integrated Online Platform for Accurate and Comprehensive Predictions for In Silico ADME/T Properties of Artemisinin and its Derivatives. Kompleksnoe Ispol′zovanie Mineral′nogo syr′â/Complex Use of Mineral Resources/Mineraldik Shikisattardy Keshendi Paidalanu 2023; 325: 14-21
• 29 Hutchings DC, Anderson SG, Caldwell JL. et al. Phosphodiesterase-5 inhibitors and the heart: compound cardioprotection?. Heart 2018; 104: 1244-1250
• 30 Kitta T, Shmohara N. PDE5 Inhibitors. Japanese Journal of Clinical Urology 2023; 70: 404-410
• 31 Kayık G, Tüzün N, Durdagi S. Investigation of PDE5/PDE6 and PDE5/PDE11 selective potent tadalafil-like PDE5 inhibitors using combination of molecular modeling approaches, molecular fingerprint-based virtual screening protocols and structure-based pharmacophore development. J Enzyme Inhib Med Chem 2017; 32: 311
• 32 Torres PHM, Sodero ACR, Jofily P. et al. Key Topics in Molecular Docking for Drug Design. International Journal of Molecular Sciences 2019; 20: 4574
• 33 Khan MS, Mohammad HA, Shahwan M. et al. Identifying Phosphodiesterase-5 Inhibitors with Drug Repurposing Approach: Implications in Vasodysfunctional Disorders. ChemistryOpen 2024; 13: e202300196
• 34 ElHady AK, El-Gamil DS, Abdel-Halim M. et al. Advancements in Phosphodiesterase 5 Inhibitors: Unveiling Present and Future Perspectives. Pharmaceuticals 2023; 16: 1266
• 35 Sethi A, Joshi K, Sasikala K. et al. Molecular Docking in Modern Drug Discovery: Principles and Recent Applications. Drug Discovery and Development – New Advances. IntechOpen 2020; 1-21
• 36 Srivastava V, Yadav A, Sarkar P. Molecular docking and ADMET study of bioactive compounds of Glycyrrhiza glabra against main protease of SARS-CoV2. Mater Today Proc 2022; 49: 2999
• 37 Perkin VO, Antonyan GV, Radchenko EV. et al. Web Services for the Prediction of ADMET Parameters Relevant to the Design of Neuroprotective Drugs. Neuromethods 2023; 203: 465-485
• 38 Carlo D, Ronchi A, De Carlo A. et al. Predicting ADMET Properties from Molecule SMILE: A Bottom-Up Approach Using Attention-Based Graph Neural Networks. Pharmaceutics 2024; 16: 776
• 39 Kloner RA, Burnett AL, Miner M. et al. Princeton IV consensus guidelines: PDE5 inhibitors and cardiac health. J Sex Med 2024; 21: 90-116
• 40 Peak TC, Richman A, Gur S. et al. The Role of PDE5 Inhibitors and the NO/cGMP Pathway in Cancer. Sex Med Rev 2016; 4: 74-84
• 41 Nhat Phuong D, Flower DR, Chattopadhyay S. et al. Towards Effective Consensus Scoring in Structure-Based Virtual Screening. Interdiscip Sci 2023; 15: 131-145

Publication History

Received: 11 June 2024

Accepted: 30 September 2024

Article published online:
12 November 2024