INTRODUCTION:

The study of drug absorption, distribution, biotransformation, and excretion in both humans and animals is termed pharmacokinetics. By using certain mathematical procedures to interpolate and extrapolate the drug concentrations, a more thorough description of the data can be obtained¹. The conversion of pharmacokinetics (PK) data from animals and the concentration of drugs in plasma against time (C_p profile) to humans is one of the main approaches for the safe and effective development of new drug molecules². The pharmaceutical industry is now targeted more toward the synthesis and exploration of naturally occurring API for various biological activities owing to their versatile properties, accessibility and fewer side effects³. Among the several classes of phytochemicals, the antioxidant and anti-inflammatory qualities of polyphenols found in a variety of botanical compounds have drawn attention⁴.

Because plants can supplement modern pharmacological procedures, conventional medicinal plant analysis has become more popular worldwide over time⁵. Research works already established the various biological properties of mangiferin (MGF) such as antioxidant, anti-infection, anticancer, anti-diabetic, cardiovascular, neuroprotective properties etc⁶. Because of their low biopharmaceutical qualities, which lead to low bioavailability, the pharmaceutical application of these substances was remained restricted⁷. This problem can be overcome by modification of the surface and releasing characteristics of the indigenous drug, by formulating it as a novel drug delivery system⁸. Since mangiferin is an indigenous medication and a BCS class IV medicine, efforts to optimize its surface and release properties may be made using a drug delivery method based on nanoparticles^9,10.

As computer science developed, in silico techniques like pharmacokinetic, pharmacodynamics and ADMET predictions were frequently employed to bring insight on the pharmacological principles behind the therapeutic effects of conventional plants¹¹. Due to the complexity of oral drug absorption, the development of predictive modelling and simulation continues to be challenging for human beings.

At the beginning stages of drug development, some usual set of PK constraints is anticipated, which include clearance (CL), Volume of distribution (V_d), fraction absorbed, rate of absorption (K_a), and subsequently bioavailability (F) for an orally administered compound¹².

Gastroplus is an in silico simulation software package based on mechanistic principles, that simulates and predicts the absorption behaviour of a drug in humans and animals. The advanced Compartmental Absorption and Transit (ACAT) model is included in Gastroplus for the prediction of absorption behaviour and the physiologically based pharmacokinetic model (PBPK) for the prediction of disposition¹³. Some minimum input parameters are needed in Gastroplus software for performing the simulation. So compound-specific input parameters such as molecular weight, lipophilicity, solubility, permeability, P_ka, unbound fraction in the plasma, blood-to-plasma concentration ratio, and clearance are well predicted by the ADMET module of the Gastroplus software. So at the beginning stage of pre-formulation, the simulation and prediction using such software can be easily performed with limited data. The continuous validation and inclusion of more data as well as an understanding of absorption, distribution, metabolism, and excretion (ADME) become available during the lifecycle of the study^12,14.

ADMET predictor is a comprehensive software platform that focuses on predicting the absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties of chemical compounds. It utilizes predictive models and algorithms based on a large database of experimental data to estimate various pharmacokinetic and toxicological parameters¹⁵. It is important to note that while ADMET Predictor provides valuable predictions, it is not a substitute for comprehensive experimental studies. Experimental validation is still necessary to confirm the predicted ADMET properties and ensure the safety and efficacy of drug candidates. Nonetheless, the software serves as a valuable tool in the beginning stages of drug discovery and development, helping researchers prioritize compounds with desirable ADMET profiles and guiding the design of optimized drug candidates¹⁶.

Recent reports suggested the significant interest of the pharmaceutical industry towards the synthesis and exploration of naturally occurring active pharmaceutical ingredients (APIs) for various biological activities owing to their versatile properties, accessibility, and fewer side effects¹⁷. Research works already established the various biological properties of mangiferin (MGF) such as antioxidant, anti-infection, anticancer, anti-diabetic, cardiovascular, neuroprotective properties, etc ¹⁸. Pharmaceutical applications of MGF are still limited due to low biopharmaceutical properties resulting in poor bioavailability¹⁹. Under these circumstances, in the early preformulation stage, Gastroplus and ADMET predictor can provide valuable insights for the formulation development of naturally occurring APIs. By utilizing both Gastroplus and ADMET predictor, researchers can benefit from a broader range of predictive models and algorithms. The combination of these software allows for a more comprehensive evaluation of compound-related properties, including physicochemical characteristics, metabolism, toxicity, and excretion. It enhances the understanding of MGFs' behavior and aids in making informed decisions during the beginning stages of drug development and formulation optimization. These insights aid in the development of effective and bioavailable formulations for naturally occurring APIs^12,14,16.

METHODS:

Predicting human plasma concentration-time profile (Cp profile) of MGF:

The software Gastroplus 9.7 version was used to predict the oral absorption and C_p profile of MGF. A database for MGF was created in the software. The default value for the different physicochemical and compound-related properties of MGF was predicted by Gastroplus and ADMET predictor was taken as initial input. Suitable gut physiology was selected with the default Absorption Scale Factor (ASF). The compartmental values were kept constant. Performed the simulation for 24h and evaluated the result. The simulated C_p profile, absorption and dissolution graph, and regional absorption profile provide a basic idea regarding the pharmacokinetic behaviour of the MGF^12,20.

Calibration of the compartmental model:

The calibration of the compartmental model was performed to validate the simulation result and also aid to develop a drug-specific absorption model²¹. Data from literature consisting of the Cp profile of MGF followed by intravenous administration in rats was selected. The graph values were converted to data by using a Webplot digitizer. The in vivo data were loaded into the software as an ipd file and drawn as a graph. The PKplus module of Gastroplus software was explored to identify the suitable compartmental model based on the R² value. The derived PK values (C_max, t_max, V_d, AUC, etc.) were loaded into the compartment tab of Gastroplus software for calibration. Then by performing the simulation, comparative results for the observed data and calculated data were obtained^20,22.

Development of a compartmental oral absorption model of MG:

The oral absorption simulation model of a drug may be developed either by using the in vivo data after oral administration either in humans or animals or in vitro dissolution data. Here, in vitro, data from the literature was taken for the same. The data were loaded in the software as a dsd file and the simulation was performed. Generate the C_p profile, regional absorption, and absorption dissolution profile^12,23–25.

Parameter Sensitivity Analysis (PSA):

The developed model for predicting the absorption behaviour of the specific drug can be used for understanding how the dosage form and or other different properties including physical and chemical properties of the drug molecule affect the predicted absorption profile of the drug. For understanding the most influential parameters that affect the oral absorption of a specific drug can be identified by using the Parameter Sensitivity Analysis (PSA) module of Gastroplus software. To provide directions toward optimizing oral exposure, the impact of certain formulation-related factors can be identified and studied by using PSA. The result of the PSA will guide the development of an optimized formulation with better therapeutic outcomes^26,27.

RESULTS AND DISCUSSION:

Currently, the prediction of intestinal absorption of drugs based on computer simulations using advanced compartmental absorption and transit (ACAT) and physiologically-based pharmacokinetic (PBPK) models is a reality²⁸. The results demonstrated the usefulness of in silico modelling tools, particularly the Gastroplus and ADMET predictor, in predicting the pharmacokinetic behaviour of MGF. The combination of these tools allowed for the evaluation of various drug-related properties and provided valuable insights into MGF's poor bioavailability, which can be critical information in the early stages of drug development and formulation optimization. The calibration of the compartmental model and development of the oral absorption model further enhanced the accuracy and applicability of the simulations. The parameter sensitivity analysis enabled the identification of key factors affecting MGF's oral absorption, facilitating the design of optimized formulations for improved therapeutic outcomes.

Predicting human plasma concentration-time profile (C_p profile) of MGF:

The structure was imported to the MGF data file. The compound-specific physicochemical properties of the drug were predicted by the Gastroplus and ADMET predictor, and values were kept as initial input values. (Table 1)

Table 1. List of MGF input parameters employed for GI simulation.

Properties	ADMET Predicted value
Molecular formula	C₁₉H₁₈O₁₁
Molecular weight	422.35
Log P	-0.59
Solubility	0.89mg/ml
PK_a	11.25
Initial dose	100mg
Dosage form	IR tablet
Mean precipitation time	900 sec
Diffusion coefficient	0.7
Drug particle density	1.7
Effective human permeability	0.27

The human fasted gut physiology was selected for the simulation. The Absorption Scale Factor (ASF) values were based on the optimum log D model SA/V value. The compartmental pharmacokinetic (PK) model was selected from the pharmacokinetic tab and performed the simulation for 24 h. The different graphs such as plasma concentration-time profile, absorption-dissolution and regional absorption graphs were obtained (Figure 1,2 and 3).

Figure 1. Predicted plasma concentration-time profile of MGF

The result of the plasma concentration-time profile of ADMET predicted the input model of MGF indicating a C_max of 0.515 μg/mL, with a t_max of 3.92 hours and 48.6% bioavailability (Table 2).

Table 2. Simulation result of PK parameter of MGF GI absorption

Parameter	Simulated value
F_a (%)	48.6
FD_p (%)	46.608
F (%)	41.766
C_max (μg/mL)	0.51523
t_max (h)	3.92
AUC _0-inf (μg. h/mL)	5.5544
AUC _0-t (μg. h/mL)	5.1385
C_max Liver (μg/mL)	0.58061

The absorption-dissolution graph revealed 100% dissolution. But the simulation result for the amount of MGF absorbed, the amount that reaches into the portal vein and systemic circulation was very less. The possibility of first-pass metabolism and poor absorption of MGF was already reported in several literatures and the simulation results comply with the poor biopharmaceutical properties of the MGF^29,30.

Figure 2.Simulation graph showing the extent of absorption and dissolution.

The absorption of MGF in different regions of the body indicated that the majority of MGF from an immediate release (IR) formulation, is absorbed in the region of the duodenum and jejunum (29.8%), while the remaining dose is absorbed in the middle and distal GI regions. So the simulated regional absorption result indicated that the first part of the intestine may be the predominant site of absorption of MGF after oral administration.

Figure 3. The regional absorption simulation result of MGF in GI

Calibration of the compartmental model:

The rat IV data from the literature was taken ³¹ and loaded as an ipd file. The PK module selected 2 compartmental models as the best-fitted compartmental model (based on the R² value)¹⁷. The pharmacokinetic parameter of the IV data was calculated by using the PK module Gastroplus software and exported. After 6 h of simulation, compare the prediction by software with observed data (Figure 4). The predicted and observed data were well-fitted with each other. The summary of the PK parameter was given in Table 3.

Table 3. Comparison of pharmacokinetic parameters between predicted and in-vivo observed data after IV administration

Parameter	Observed	Simulated
Fa (%):	0	99.829
FDp (%):	0	99.826
F (%):	0	99.826
C_max (μg/ml):	13.5	13.343
t_max (h):	0.01	0
AUC _0-inf (μg.h/ml):	20.004	18.874
AUC _0-t (μg.h/ml):	19.805	17.822
C_max Liver (μg/ml):		12.303

Figure 4. Gastro Plus simulated and observed mean MGF plasma concentration vs time graph following administration of a 15mg IV bolus of MGF in the rat.

Development of a compartmental oral absorption model of MGF:

The Cp data after oral administration of MGF 0.3gm was taken from the literature³² The data was loaded as an opd file, and PK parameters from the best-fitted compartment model (2-compartment model) were exported to the corresponding tab, then performed the simulation for 24hours²⁰. The result showed that the predicted and observed data were well-fitted with each other indicating the development of a validated absorption model for MGF (Fig No.5). The summary of PK parameters of both observed and predicted was given in Table 4.

Table 4: Comparison in data between pharmacokinetic parameters of simulated and observed results.

Parameter	Observed	Simulated
Fa (%):	0	22.75
FDp (%):	0	22.21
F (%):	0	19.90
C_max (μg/mL):	26.55	26.05
t_max (h):	4.51	4.8
AUC _0-inf (μg.h/mL):	292.4	268.18
AUC _0-t (μg.h/mL):	257.53	251.84
C_max Liver (μg/mL):		26.42

Figure 5. Gastro Plus simulated and observed mean MGF plasma concentration vs time graph following administration of a 0.3gm MGF dose from IR formulation.

Parameter Sensitivity Analysis (PSA):

PSA can aid to identify the most critical physiological, physicochemical and or formulation parameters that influence absorption and oral bioavailability. Once the critical factors are identified, it may be easy to opt for a strategy that enables the formulation to overcome these limitations. Such strategies include the micronisation of particle size, the addition of solubilizing agent, co-solvent, permeability or penetration enhancer etc. In such a way, the formulator can protect excessive periods, and effort and reduce the loss of resources in the preformulation process. The PSA module of Gastroplus software enables the formulator the understanding the effect of different physiological conditions, physicochemical properties as well as formulation-related factors^14,26,27.

The PSA result indicated that solubility, permeability and initial dose will have the greatest influence on MGF oral absorption (Fig No.6). If the solubility increases from 0.086mg/ml to 0.86, then absorption increases from 5% to 45%. Further, an increase in solubility does not have much more effect on oral bioavailability. The optimum dose for maximum MGF absorption may be less than 900mg. A dose of more than 900mg may decrease the absorption.

The most important factor that affects oral absorption of MGF is effective permeability (Peff). The PSA result indicated that Peff is a direct effect on the oral bioavailability of the MGF. From the graph, it is evident that, when Peff increases from 0.3cm/s to 1.5cm/s, the % bioavailability is increased from 20 to 80%. This may be a critical point in the development of a novel delivery system with improved oral absorption of MGF.

Figure 6. PSA result 0.3mg oral MGF

CONCLUSION:

The in silico modeling tools Gastroplus and ADMET Predictor were employed to predict the oral absorption and Cp profile of MGF, a naturally occurring active pharmaceutical ingredient (API) with various biological properties. The simulation results demonstrated that MGF possesses poor bioavailability due to its suboptimal biopharmaceutical properties and provided valuable insights into its pharmacokinetic behavior. The Gastroplus software's physiologically based pharmacokinetic (PBPK) model and advanced compartmental absorption and transit (ACAT) model enabled the prediction of drug absorption behavior and disposition in both humans and animals. Conversely, the ADMET Predictor focused on forecasting various pharmacokinetic and toxicological parameters using a substantial experimental database. These predictive methods were useful in the early stages of drug development and formulation optimization, assisting researchers in prioritizing compounds with favorable ADMET profiles and creating improved drug candidates. The study's primary objectives included establishing a database for MGF, predicting compound-specific features, simulating oral absorption, calibrating the compartmental model, and developing an oral absorption model based on in vitro data. Parameter sensitivity analysis (PSA) was also performed in the study to identify crucial variables affecting the oral absorption and bioavailability of MGF. The results indicated that solubility, permeability, and initial dose were significant factors influencing MGF's oral absorption, with effective permeability playing a critical role in enhancing bioavailability. The findings may be utilized to create more potent and bioavailable versions of naturally occurring APIs like MGF, thus expanding their therapeutic applications. However, the study's major limitations include the need for further experimental validation to confirm the predicted ADMET features, safety, and efficacy of MGF or any other drug molecule. Additionally, Gastroplus and ADMET Predictor may not fully account for the intricate processes of the GI system, such as efflux transporters or gut wall metabolism.

ACKNOWLEDGEMENT:

The authors would like to thank CPS, GMC Kannur, Kerala for providing the software required for the study.

CONFLICT OF INTEREST:

The authors declare that there is no conflict of interest.

ABBREVIATIONS:

API: Active Pharmaceutical Ingredient; MGF: Mangiferin; Cp: PK: Pharmacokinetics; Plasma Concentration; PSA: Parameter Sensitivity Analysis: Peff: Effective Permeability; IV: Intravenous; PBPK: Physiologically Based Pharmacokinetic; ACAT: Advanced Compartmental Absorption and Transit

REFERENCE:

1. Debnath S. Kumar T. H. H. An Overview on Pharmacokinetics and Pharmacokinetic Modeling. Asian J. Res. Pharm. Sci. 2020; 10, 124.

2. Hu Y. Smith D. E. In Silico Prediction of the Absorption and Disposition of Cefadroxil in Humans using an Intestinal Permeability Method Scaled from Humanized PepT1 Mice. Drug Metab. Dispos. 2019; 47, 173–183.

3. Arivukkarasu R. Rajasekaran, A. Hussain S. H. B. Ajnas M. Simultaneous detection of Rutin, Quercetin, Gallic acid, Caffeic acid, Ferulic acid, Coumarin, Mangiferin and Catechin in Hepatoprotective commercial herbal formulations by HPTLC technique. Res. J. Pharmacogn. Phytochem. 2018; 10: 59.

4. Abriyani E. Fikayuniar L. Screening phytochemical, antioxidant activity and vitamin c assay from bungo perak-perak (Begonia versicolar irmsch) leaves. Asian J. Pharm. Res. 2020; 10: 183.

5. Singh A. A Review of various aspects of the Ethnopharmacological, Phytochemical, Pharmacognostical, and Clinical significance of selected Medicinal plants. Asian J. Pharm. Technol. 2022; 349–360 doi:10.52711/2231-5713.2022.00055.

6. Baseer Khan AJ B. Arif M. Microwave Supported Extraction and Optimization of Flavonoid Mangiferin from Mangifera indica L. Stem Bark using Orthogonal Array Design. Res. J. Pharm. Technol. 2023; 1113–1117 doi:10.52711/0974-360X.2023.00185.

7. B Mathew J. Fathima Z. Raviraj C. Mathew A. Quantitative Estimation of Mangiferin and Molecular Docking Simulation of Salacia reticulata Formulation. Res. J. Pharm. Technol. 2024; 578–584 doi:10.52711/0974-360X.2024.00090.

8. Indalkar YR. Pimpodkar NV. Godase AS. Gaikwad PS. A Compressive Review on the Study of Nanotechnology for Herbal Drugs. Asian J. Pharm. Res. 2015; 5: 203.

9. Nistane NT. Herbal Nanoparticles against Cancer. Res. J. Pharm. Dos. Forms Technol. 2019; 11: 247.

10. Ahmad W.Khan T. Basit, I. Imran JA Comprehensive Review on Targeted Drug Delivery System. Asian J. Pharm. Res.2022; 335–340 doi:10.52711/2231-5691.2022.00053.

11. Samajdar S. In silico ADME prediction of Phytochemicals present in Piper longum. Asian J. Res. Pharm. Sci.2023; 89–93 doi:10.52711/2231-5659.2023.00017.

12. Grbic S. Parojcic J. Djuric Z. Computer-aided biopharmaceutical characterization: gastrointestinal absorption simulation. in Computer-Aided Applications in Pharmaceutical Technology. Elsevier. 2013; 177–232. doi:10.1533/9781908818324.177.

13. Habeeb, M. et al. Development Characterization and Molecular Simulation studiesof Metoclopramide HCl and Tramadol HCl Bilayer Tablets. Res. J. Pharm. Technol. 2022; 529–534. doi:10.52711/0974-360X.2022.00085.

14. Hosea NA.Jones HM. Predicting Pharmacokinetic Profiles Using in Silico Derived Parameters. Mol. Pharm. 2013; 10: 1207–1215.

15. M Mukadam. MM Jagdale D. In silico ADME/T Prediction of Steroidal Chalcone derivatives using Swiss ADME and OSIRIS explorer. Res. J. Pharm. Technol. 2024; 843–848 doi:10.52711/0974-360X.2024.00130.

16. Sohlenius-Sternbeck AK. Terelius Y. Evaluation of ADMET Predictor in Early Discovery Drug Metabolism and Pharmacokinetics Project Work. Drug Metab. Dispos. 2022; 50, 95–104.

17. Nitin D. Love Kumar S. Prediction of In-silico ADMET Properties and Molecular docking study of Substituted Thiadiazole for screening of Antiviral activity against protein target Covid-19 main protease. Res. J. Pharm. Technol. 2023; 5802–5807. doi:10.52711/0974-360X.2023.00939.

18. PubChem. Mangiferin. https://pubchem.ncbi.nlm.nih.gov/compound/5281647.

19. Morozkina SN. Nhung Vu TH. Generalova YE. Snetkov PP. Uspenskaya MV. Mangiferin as New Potential Anti-Cancer Agent and Mangiferin-Integrated Polymer Systems—A Novel Research Direction. Biomolecules 2021; 11, 79.

20. Duque MD. et al. In Silico Prediction of Plasma Concentrations of Fluconazole Capsules with Different Dissolution Profiles and Bioequivalence Study Using Population Simulation. Pharmaceutics. 2019; 11: 215.

21. Suharti N. et al. In Silico Prediction and In Vitro Cytotoxic Activity of Arbuscular Mycorrhizal Fungi Induced Zingiber officinale Var. Rubrum. Res. J. Pharm. Technol. 2022; 4913–4918, doi:10.52711/0974-360X.2022.00825.

22. Vardhan H. Mittal P. Adena SKR.Upadhyay M. Mishra B. Development of long-circulating docetaxel loaded poly (3-hydroxybutyrate-co-3-hydroxyvalerate) nanoparticles: Optimization, pharmacokinetic, cytotoxicity and in vivo assessments. Int. J. Biol. Macromol. 2017; 103: 791–801.

23. Sjögren E. Thörn H. Tannergren C. In Silico Modeling of Gastrointestinal Drug Absorption: Predictive Performance of Three Physiologically Based Absorption Models. Mol. Pharm. 2016; 13: 1763–1778.

24. Sjögren, E. et al. In silico predictions of gastrointestinal drug absorption in pharmaceutical product development: Application of the mechanistic absorption model GI-Sim. Eur. J. Pharm. Sci. 2013; 49: 679–698.

25. Grbic S. Lukic V. Kovacevic I. Parojcic J. Djuric Z. An investigation into the possibilities and limitations of in silico absorption modeling: GastroPlusTM simulation of nimesulide oral absorption. in Proceedings of The 2nd Electronic Conference on Pharmaceutical Sciences 816 (MDPI, Sciforum.net, 2012). doi:10.3390/ecps2012-00816.

26. Grbic S. Lukic V. Kovacevic I. Parojcic J. Djuric Z. An investigation into the possibilities and limitations of in silico absorption modeling: GastroPlusTM simulation of nimesulide oral absorption. in Proceedings of The 2nd Electronic Conference on Pharmaceutical Sciences 816 (MDPI, Sciforum.net, 2012). doi:10.3390/ecps2012-00816.

27. Kovacevic I. Parojčić J. Homšek I. Tubić-Grozdanis M. Langguth P. Justification of Biowaiver for Carbamazepine, a Low Soluble High Permeable Compound, in Solid Dosage Forms Based on IVIVC and Gastrointestinal Simulation. Mol. Pharm. 2009; 6: 40–47.

28. Peter JS. et al. In Silico approach to predict the potential binding affinity of the active Ingredient of the Macrotyloma uniflorum seed against orphan nuclear receptor. Res. J. Pharm. Technol. 2021; 14: 694–700.

29. Prado Y. et al. Acute and 28-day subchronic toxicity studies of mangiferin, a glucosyl xanthone isolated from Mangifera indica L. stem bark. 11.

30. Ma H. Chen H. Sun L. Tong L. Zhang T. Improving permeability and oral absorption of mangiferin by phospholipid complexation. Fitoterapia 2014; 93: 54–61.

31. Kammalla AK. et al. Comparative Pharmacokinetic Study of Mangiferin After Oral Administration of Pure Mangiferin and US Patented Polyherbal Formulation to Rats. AAPS PharmSciTech 2015; 16: 250–258.

32. Khurana RK. et al. Exploring and validating physicochemical properties of mangiferin through GastroPlus® software. Future Sci. 2017; OA 3: FSO167.