INTRODUCTION:

Diabetes mellitus is a complex, chronic metabolic disorder characterized by sustained hyperglycemia, resulting from either insufficient insulin production (Type 1 diabetes) or insulin resistance in peripheral tissues (Type 2 diabetes)¹. It has emerged as a global health concern, affecting over 537 million adults as of 2021, with the numbers projected to rise to 783 million by 2045². The disease is associated with severe complications, including cardiovascular diseases, nephropathy, neuropathy, and retinopathy, which significantly affect morbidity and mortality rates³. Effective management of diabetes requires a combination of lifestyle changes and pharmacological intervention.

Antidiabetic agents play a pivotal role in glycemic control and reducing the risk of diabetes-related complications. The therapeutic arsenal for diabetes management includes a wide range of drug classes such as biguanides (e.g., metformin), sulfonylureas (e.g., glimepiride), thiazolidinediones (e.g., pioglitazone), glucagon-like peptide-1 receptor agonists (e.g., liraglutide), and sodium-glucose co-transporter-2 inhibitors (e.g., dapagliflozin)⁴. Despite the diversity of therapeutic options, the quest for safer and more effective drugs remains a priority in diabetes research.

While currently marketed antidiabetic agents have improved diabetes management, they are not devoid of limitations. Metformin, the first-line treatment, is associated with gastrointestinal side effects and risks of lactic acidosis in certain patients⁵. Sulfonylureas, although effective, can lead to hypoglycemia and weight gain⁶. Other classes, such as thiazolidinediones, have been linked to fluid retention, heart failure, and bone fractures⁷. Poor oral bioavailability, rapid metabolism, and off-target effects further limit the clinical efficacy of many antidiabetic drugs. These challenges underscore the need for ongoing optimization to enhance safety and therapeutic outcomes.

Advances in computational tools have revolutionized drug discovery and development, providing efficient and cost-effective strategies to overcome the limitations of traditional experimental methods. In silico approaches allow for the rapid evaluation of absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties of drug candidates, enabling the identification of compounds with optimal pharmacokinetic and pharmacodynamic profiles⁸. Tools such as SwissADME, pkCSM, and ProTox-II facilitate the prediction of bioavailability, synthetic accessibility, and toxicity, offering valuable insights into drug-likeness and safety profiles^9,10. Computational methods also enable high-throughput screening of drug candidates, thereby reducing the time and cost associated with experimental validation.

This study aims to analyze the ADME, toxicity, bioavailability, and synthetic accessibility profiles of selected marketed antidiabetic agents using computational approaches. By leveraging in silico tools, the research seeks to identify compounds with favorable pharmacokinetic properties and minimal toxicity, providing a foundation for the development of optimized therapeutic agents for diabetes management.

MATERIALS AND METHODS:

Selection of Antidiabetic Agents:

Antidiabetic agents were selected based on their clinical relevance, common usage, and availability. The data were extracted from a comprehensive literature survey. The selected agents represent a range of drug classes, including biguanides, sulfonylureas, DPP-4 inhibitors, SGLT2 inhibitors, GLP-1 receptor agonists, and insulin analogs.

Preparation of Antidiabetic Agent Structures:

The molecular structures of the selected antidiabetic agents were prepared using ChemDraw software. This step ensured that all functional groups and stereochemistry were accurately represented. Then, the structures were extracted in the SMILES format.

In Silico Screening:

Drug Likeness and ADMET Analysis:

The drug-likeness and ADMET profiles were analyzed to determine the suitability of the compounds for therapeutic use. Lipinski’s Rule of Five: Each compound was evaluated based on molecular weight, hydrogen bond donors and acceptors, and lipophilicity (logP). Compounds violating these rules were flagged for potential issues with bioavailability. ADMET Analysis: Comprehensive ADMET assessments were conducted using pkCSM and SwissADME. This included predictions for gastrointestinal absorption, CYP450 enzyme interactions, and potential toxicological concerns such as hepatotoxicity and mutagenicity¹¹.

In Silico Toxicity Predictions:

ProTox 3.0 server based online tool was used to predict potential toxicity parameters like Hepatotoxicity, Cardiotoxicity, Genotoxicity and Mutagenicity^12,13.

RESULTS AND DISCUSSION:

The selected antidiabetic agents for the study include representatives from various pharmacological classes commonly used in the management of diabetes mellitus. From the Biguanides class, Metformin was selected due to its widespread use and well-established efficacy. Among the Sulphonylureas, agents such as Tolbutamide, Chlorpropamide, Glipizide, and Glimipiride were included for their insulin secretagogue activity. Dipeptidyl Peptidase-4 (DPP-4) inhibitors like Sitagliptin and Tenegliptin were chosen for their role in enhancing incretin activity. Thiazolidinediones, represented by Pioglitazone and Rosiglitazone, were selected for their insulin-sensitizing effects. From the class of Glucosidase inhibitors, Acarbose and Voglibose were included due to their ability to delay carbohydrate absorption. Sodium-Glucose Cotransporter-2 (SGLT2) inhibitors such as Dapagliflozin, Canagliflozin, and Bexagliflozin were selected for their novel mechanism of increasing urinary glucose excretion. Lastly, the Meglitinides, Repaglinide and Nateglinide, were chosen for their short-acting insulinotropic effects. These agents were selected to represent a broad spectrum of mechanisms and therapeutic strategies in antidiabetic treatment.

Structural Representation of Selected Antidiabetic Agents:

ChemDraw software was used for the preparation and visualization of chemical structures of antidiabetic agents (Figure 1). This tool facilitated the accurate representation of molecular structures.

Figure 1: Structures of Selected Antidibetic Agents

Drug Likeness and ADMET Analysis:

The Drug likeness characteristics and ADME studies (Table 1 and 2) of selected antidibetic agents were evaluated by using SwissADME online Programme. Lipinski's Rule of Five is a set of guidelines that predicts the drug-likeness of a compound. The "Rule of Five" gets its name because each rule involves a number that is a multiple of five. Lipinski's Rule of Five states that an orally active drug should have molecular weight < 500, logP < 5, less than 5 hydrogen bond donors, and less than 10 hydrogen bond acceptors.

Table 1: Lipinski’s Rule of Five Analysis for Selected Antidiabetic Agents

Antidibetic Agent	MW (g/mol)	mLOGP	nHBA	nHBD	MR (Molar Refractivity)
Metformin	129.16	-0.56	2	3	36.93
Tolbutamide	270.35	1.62	3	2	69.92
Glimepiride	490.62	2.28	5	3	133.31
Glipizide	445.54	0.62	6	3	115.30
Chlorpropamide	276.74	1.60	3	2	65.16
Sitagliptin	407.31	2.93	10	1	87.25
Tenegliptin	426.58	1.86	4	1	135.53
Pioglitazone	356.44	2.01	4	1	102.17
Rosiglitazone	357.43	1.64	4	1	101.63
Acarbose	645.60	-6.94	19	14	136.69
Voglibose	267.28	-3.44	8	8	59.04
Dapagliflozin	408.87	1.07	6	4	104.82
Canagliflozin	444.52	1.95	6	4	116.75
Repaglinide	452.59	3.44	4	2	135.45
Nateglinide	317.42	2.81	3	2	91.75
Bexagliflozin	464.94	0.91	7	4	118.22

1) Lipinski’s Rule of Five: Among the listed antidiabetic agent, Acarbose demonstrates a violation of Lipinski’s rule of five, particularly due to the number of hydrogen bond acceptors (greater than 10), hydrogen bond donors (greater than 5), and a molecular weight (MW) exceeding 500 Daltons. These factors collectively suggest poor oral bioavailability, indicating that Acarbose may have limited absorption when administered via the oral route.

While, Voglibose violates Lipinski’s rule due to its high number of hydrogen bond donors, which exceed the recommended threshold of 5. This deviation may affect its pharmacokinetic profile, particularly its absorption characteristics.

2) Partition Coefficient: Calculating logP, which measures how a compound divides between water and fat, is important in drug design. Various methods like WLOGP, XLOGP, and iLOGP are used to predict logP, but each has its pros and cons. WLOGP and XLOGP, which are fragment-based methods, often overestimate logP, particularly for large molecules. The atomistic approach created by Wildman and Crippen30 is implemented by us under the name WLOGP. An internal application of Moriguchi's topological approach is used to calculate MLOGP.

Drug absorption depends on a compound's ability to pass through cell membranes, which is measured by LogP. Implicit log P, or iLOGP, is an internal physics-based technique that uses GB/SA's calculation of the Gibbs free energy of solvation in water and n-octanol. Good bioavailability, balanced solubility, and appropriate distribution in tissues are all ensured by an optimal logP. It is critical for safety since high logP can accumulate in fatty tissues and possibly cause toxicity. For drug-like properties, an optimal logP range of 1 to 3 is ideal for most orally active drugs, as it balances solubility and permeability, allowing the drug to be effectively absorbed and distributed in the body.

a) LOGP < 1: Suggests that the molecule is more hydrophilic, which could lead to poor membrane permeability.

b) LOGP 1-3: Ideal for oral drugs, balancing solubility and permeability.

c) LOGP > 3: Suggests that the molecule is increasingly lipophilic, which could improve permeability but may lead to solubility and bioavailability issues.

d) LOGP > 5: Compounds with such high values may face bioavailability issues due to excessive lipophilicity, and they may also be prone to toxicity due to accumulation in tissue.

Drugs like Acarbose, Voglibose and Metformin having logP value is less than 1, it means the drugs are too hydrophilic, among these all three agents Acarbose is too hydrophilic which having logP value is -8.53 which having BA score is very less, while the drugs like Repaglinide and Nateglinide which having logP value is near about 5, means the drugs are too lipophilic.

3) Topological Polar Surface Area (TPSA): The TPSA should ideally be less than 140 Å² to ensure efficient absorption and distribution, particularly for oral drugs. A high TPSA is often associated with poor membrane permeability, which can hinder absorption through the intestinal lining and reduce the compound's ability to cross barriers such as the blood-brain barrier.

As per results, agents such as Acarbose and Voglibose exhibit TPSA values that exceed this threshold, which may limit their bioavailability.

4) Rotatable Bonds: Compounds with a high number of rotatable bonds (typically more than 9) tend to have reduced oral bioavailability and poorer pharmacokinetic properties. This is due to increased molecular flexibility, which negatively impacts membrane permeability and metabolic stability. In the analysis, agents such as Glimepiride and Repaglinide show violations in this aspect, which could lead to decreased permeability and bioavailability.

5) Synthetic Accessibility: Synthetic accessibility is quantified on a scale from 1 to 10, where 1 indicates compounds that are easy to synthesize and 10 represents compounds that are more complex to produce. According to the data, agents such as Metformin, Glipizide, Sitagliptin, Pioglitazone, Voglibose, Rosiglitazone, Repaglinide, and Nateglinide have synthetic accessibility scores closer to 1, indicating that these compounds are relatively easy to synthesize. Other compounds like Glimepiride, Bexagliflozin, Dapagliflozin, Canagliflozin, and Teneligliptin have moderate synthetic accessibility scores, near 5, suggesting that their synthesis involves moderate complexity. On the other hand, Acarbose has the highest synthetic accessibility score (7.34) among the listed agents, indicating that its synthesis process is considerably more complex.

Table 2: Pharmacokinetic and Drug-likeness Parameters of Antidiabetic Agents

Antidiabetic Agent	TPSA	iLOGP	XLOGP3	WLOGP	nROT	Lipinski Violations	Veber Violations	Synthetic Accessibility
Metformin	91.49	0.34	-1.27	-1.24	2	0	0	3.02
Tolbutamide	83.65	1.54	2.34	2.86	7	0	0	2.42
Glimepiride	133.06	2.42	3.85	3.77	11	0	1	4.71
Glipizide	138.53	2.51	1.91	3.16	10	0	0	3.33
Chlorpropamide	83.65	1.28	2.27	2.82	6	0	0	2.37
Sitagliptin	77.04	2.35	0.70	3.90	6	0	0	3.50
Tenegliptin	81.94	3.30	2.41	0.04	5	0	0	4.30
Pioglitazone	93.59	2.61	3.75	2.78	7	0	0	3.46
Rosiglitazone	96.83	2.41	3.11	2.11	7	0	0	3.35
Acarbose	321.17	-0.35	-8.53	-8.56	9	3	1	7.34
Voglibose	153.64	0.88	-4.09	-4.49	5	1	1	3.66
Dapagliflozin	99.38	3.12	2.35	1.52	6	0	0	4.52
Canagliflozin	118.39	3.27	3.23	3.06	5	0	0	4.99
Repaglinide	78.87	3.81	5.18	4.51	11	0	1	3.89
Nateglinide	66.40	2.52	4.06	3.26	7	0	0	3.22
Bexagliflozin	108.61	3.99	2.39	1.62	9	0	0	4.90

Table 2: ADME Analysis of Selected Antidibetic Agents

Antidiabetic Agent	GI Absorption	BBB Permeant	Pgp Substrate	CYP1A2 Inhibitor	CYP2C19 Inhibitor	CYP2C9 Inhibitor	CYP2D6 Inhibitor	CYP3A4 Inhibitor	Bioavailability Score	PAINS Alerts
Metformin	High	No	No	No	No	No	No	No	0.55	0
Tolbutamide	High	No	No	No	No	No	No	No	0.55	0
Glimepiride	Low	No	Yes	No	No	Yes	No	Yes	0.55	0
Glipizide	Low	No	Yes	No	No	Yes	Yes	Yes	0.55	0
Chlorpropamide	High	No	No	No	No	No	No	No	0.55	0
Sitagliptin	High	Yes	Yes	No	No	No	No	No	0.55	0
Tenegliptin	High	No	No	No	No	No	Yes	Yes	0.55	0
Pioglitazone	High	No	No	Yes	Yes	Yes	Yes	Yes	0.55	0
Rasiglitazone	High	No	No	No	Yes	Yes	Yes	Yes	0.55	0
Acarbose	Low	No	Yes	No	No	No	No	No	0.17	0
Voglibose	Low	No	Yes	No	No	No	No	No	0.55	0
Dapagliflozin	High	No	Yes	No	No	No	Yes	No	0.55	0
Canagliflozin	High	No	Yes	Yes	No	No	Yes	No	0.55	0
Repaglinide	High	No	Yes	No	No	Yes	Yes	Yes	0.56	0
Nateglinides	High	Yes	No	No	No	Yes	No	No	0.85	0
Bexagliflozin	High	No	Yes	No	No	No	Yes

ADME and Bioavailability Study:

1) GI absorption and Bioavailability Score: Among the drugs discussed in Table 2, certain antidiabetic agents exhibit varying levels of gastrointestinal (GI) absorption and bioavailability scores. The drugs which having high permeability and solubility ultimately shows high GI absorption and bioavailability. Specifically, drugs such as Glimipiride, Glipizide, Acarbose, and Voglibose demonstrate low GI absorption when administered orally. The bioavailability scores for these medications range from 0.17 to 0.85, indicating that only 17% to 85% of the administered dose reaches systemic circulation. Among these, Acarbose presents a particularly low bioavailability score of 0.17, attributed to its limited GI absorption. This low bioavailability suggests that alternative routes of administration should be considered for Acarbose to enhance its systemic availability. Conversely, Nateglinide exhibits high GI absorption, resulting in a significantly higher bioavailability score of 0.85, meaning that 85% of the drug is available in systemic circulation following oral administration. This favorable absorption profile makes Nateglinide a more effective option in terms of achieving therapeutic concentrations in the bloodstream.

2) BBB Permeant: BBB permeant refers to a substance that has the ability to cross the blood-brain barrier (BBB), a highly selective barrier that protects the brain from potentially harmful chemicals while allowing essential nutrients to pass through. The BBB is composed of tightly packed endothelial cells, astrocytes, and a basement membrane that regulate the passage of substances between the bloodstream and the central nervous system (CNS). Among the above list almost all drugs are impermeable to BBB but the drugs like Sitagliptin and Nateglinides are permeable to BBB.

3) Pgp Substrate: P-glycoprotein (P-gp), also known as multidrug resistance protein 1 (MDR1) or ABCB1, is a member of the ATP-binding cassette (ABC) transporter family. It plays a significant role in the efflux of a wide variety of drugs and xenobiotics across cellular membranes, thereby influencing drug absorption, distribution, and excretion. P-gp is highly expressed in the intestinal epithelium, where it limits the absorption of various orally administered drugs by pumping them back into the gut lumen. This reduces the bioavailability of drugs, meaning less of the drug reaches systemic circulation, thus affecting therapeutic efficacy. A P-glycoprotein (P-gp) substrate is a compound that is recognized and transported by the P-glycoprotein efflux pump.When a drug is a P-gp substrate, it means that P-gp can reduce its intracellular concentration by transporting it out of the cell, which has significant implications for the drug's pharmacokinetics (absorption, distribution, metabolism, and excretion). In the above mention list in (table1.2) the drugs like Glimipiride, Glipizide, Acarbose, Voglibose, Sitagliptin, Dapagliflozin, Canagliflozin, and Boxagliflozin are pgp substrates so bioavailability of these drugs somewhat gets reduced due to efflux by pgp substrate.

4) Cytochrome P450 Inhibitors: Cytochrome P450 is a crucial detoxification enzyme in the body, primarily located in the liver. It oxidises xenobiotics to aid their elimination. Many drugs are deactivated by P450s, while others can be activated by them. It is therefore important to assess a compound's ability to inhibit cytochrome P450. Models for various isoforms (CYP1A2/CYP2C19/ CYP2C9/CYP2D6/CYP3A4) were created using over 14000 to 18000 compounds whose ability to inhibit cytochrome P450 was determined. Cytochrome P450s are responsible for the metabolism of many drugs. However, P450 inhibitors can significantly alter the pharmacokinetics of these drugs. It is therefore necessary to determine whether a given compound is likely to be a cytochrome P450 substrate.

In the above mention list of antidiabetics agents, the agents like Glimipiride, Glipizide, Tenegliptin, Pioglitazone, Rasiglitazone, Canagliflozine, Dapagliflozin, Bexagliflozin, Nateglinide and repaglinide mainly inhibits the cytochromep450 enzymes, among which Glimipiride mainly inhibit the enzyme CYP2C9and CYP3A4, Glipizide and Repaglinide inhibits the enzyme like CYP2C9, CYP2D6, CYP3A4 then agents like Bexagliflozin, Canagliflozin and Dapagliflozin mainly inhibits the enzyme CYP2D6. Pioglitazone mainly inhibits all the enzymes like CYP1A2, CYP2C19, CYP2C9, CYP2D6, CYP3A4, this drug slowing down the metabolism, this can lead to increased blood levels of these drugs, potentially causing toxicity or enhanced pharmacological effects.

The drugs like Metformin, Chlorpropamide, Sitagliptin, Acarbose, Voglibose, Tolbutamide cannot inhibit ant type of enzyme so it can be metabolized easily from the body, so the chances of toxicity should be minimum in case of these agents. The agent Pioglitazone mainly inhibits the all four types of enzymes which having chances of toxicity due to reduction in metabolism rate.

5) PAINS Alerts: PAINS (Pan-Assay Interference Compounds) are molecules that can cause misleading results in biological tests. These compounds often show strong activity in various assays, regardless of the specific protein they are meant to target. The importance of PAINS alerts in drug discovery is significant, as they help researchers identify and avoid misleading compounds that could complicate the development of effective drugs in simpler terms, PAINS are "false alarms" in drug discovery that can lead to the mistaken belief that a compound is effective when it may just be reacting non-specifically in different tests. PAINS alerts help distinguish true drug candidates from compounds that show non-specific activity in assays, thereby reducing the occurrence of false positives. This leads to more reliable screening results. Therefore, all the drugs listed above having zero PAINS alerts which cannot be show misleading results.

In Silico Toxicity Predictions:

There are many tools and databases available for toxicity studies which mainly include Derek Nexus, ToxTree, Protox 3.0. Among this all we used Protox 3.0 for this study. With the help of the free online tool ProTox 3.0, a user can quickly determine the toxicity of an input substance. Predictions of computational toxicity have the potential to minimize the number of animal tests conducted and preserve animal life. For a variety of toxicity endpoints, ProTox 3.0 integrates machine-learning models with chemical similarity. A unique feature of the ProTox webserver is the classification of the prediction scheme into levels of toxicity, such as acute rodent toxicity (oral toxicity), hepatotoxicity (organ toxicity) and toxicological endpoints (mutagenicity, carcinotoxicity, cytotoxicity, and toxicity). This classification offers valuable information about potential molecular mechanisms underlying toxic responses.

Toxic doses are often given as LD50 values in mg/kg body weight. The LD50 is the median lethal dose meaning the dose at which 50% of test subjects die upon exposure to a compound. Toxicity classes are defined according to the globally harmonized system of classification of labelling of chemicals (GHS). LD50 values are given in [mg/kg]:

Class I: Fatal if swallowed (LD50 ≤ 5)

Class II: Fatal if swallowed (5 < LD50 ≤ 50)

Class III: Toxic if swallowed (50 < LD50 ≤ 300)

Class IV: Harmful if swallowed (300 < LD50 ≤ 2000)

Class V: May be harmful if swallowed (2000 < LD50 ≤ 5000)

Class VI: Non-toxic (LD50 > 5000)

Table 3: Toxicity Analysis of Selected Antidibetic Agents

Antidiabetic Agents	Predicted LD50	Predicted Toxicity Class	Organ Toxicity
Antidiabetic Agents	Predicted LD50	Predicted Toxicity Class	Hepatotoxicity	Neurotoxicity	Nephrotoxicity	Respiratory Toxicity	Cardiotoxicity
Metformin	680 mg/kg	4	Inactive	Inactive	Inactive	Inactive	Inactive
Tolbutamide	490 mg/kg	4	Inactive	Inactive	Active	Active	Inactive
Glimepiride	4000 mg/kg	5	Inactive	Inactive	Active	Active	Inactive
Glipizide	15000 mg/kg	6	Inactive	Inactive	Active	Active	Inactive
Chlorpropamide	1100 mg/kg	4	Inactive	Inactive	Active	Active	Inactive
Sitagliptin	2500 mg/kg	5	Inactive	Active	Inactive	Active	Inactive
Tenegliptin	1300 mg/kg	4	Inactive	Active	Inactive	Active	Inactive
Pioglitazone	1000 mg/kg	4	Active	Active	Inactive	Active	Inactive
Rasiglitazone	1000 mg/kg	4	Inactive	Active	Inactive	Active	Inactive
Acarbose	24000 mg/kg	6	Active	Inactive	Active	Active	Active
Voglibose	14700 mg/kg	6	Inactive	Inactive	Active	Active	Active
Dapagliflozin	3000 mg/kg	5	Inactive	Inactive	Active	Active	Active
Canagliflozin	2160 mg/kg	5	Inactive	Inactive	Active	Active	Active
Repaglinide	1420 mg/kg	4	Inactive	Active	Active	Active	Inactive
Nateglinides	800 mg/kg	4	Inactive	Inactive	Inactive	Active	Active
Bexagliflozin	3000 mg/kg	5	Inactive	Inactive	Active	Active	Active

Table 4: Toxicity End Points Analysis of Selected Antidibetic Agents

Antidiabetic Agents	Toxicity End Points
Antidiabetic Agents	Carcinogenicity	Immunotoxicity	Mutagenicity	Cytotoxicity	BBB Barrier Permeability
Metformin	Inactive	Inactive	Inactive	Inactive	Active
Tolbutamide	Inactive	Inactive	Inactive	Inactive	Active
Glimepiride	Inactive	Inactive	Inactive	Inactive	Active
Glipizide	Inactive	Inactive	Inactive	Inactive	Active
Chlorpropamide	Inactive	Inactive	Inactive	Inactive	Active
Sitagliptin	Inactive	Inactive	Inactive	Inactive	Active
Tenegliptin	Active	Inactive	Inactive	Inactive	Active
Pioglitazone	Inactive	Inactive	Inactive	Inactive	Active
Rasiglitazone	Active	Inactive	Inactive	Inactive	Active
Acarbose	Inactive	Active	Inactive	Inactive	Inactive
Voglibose	Inactive	Active	Inactive	Inactive	Inactive
Dapagliflozin	Inactive	Inactive	Inactive	Inactive	Active
Canagliflozin	Inactive	Inactive	Inactive	Inactive	Active
Repaglinide	Inactive	Inactive	Inactive	Inactive	Inactive
Nateglinides	Inactive	Inactive	Inactive	Inactive	Active
Bexagliflozin	Inactive	Inactive	Inactive	Inactive	Active

The organ toxicities of selected antidiabetic agents were predicted using the ProTox 3.0 web server, with a focus on hepatotoxicity, neurotoxicity, nephrotoxicity, respiratory toxicity, and cardiotoxicity. The following observations were made:

1) Hepatotoxicity: Pioglitazone and Acarbose were predicted to exhibit active hepatotoxicity, while the remaining agents showed inactive hepatotoxicity.

2) Neurotoxicity: Sitagliptin, Tenegliptin, Pioglitazone, Rasiglitazone, and Repaglinide demonstrated active neurotoxicity. All of these agents, except Repaglinide, were found to be inactive for nephrotoxicity.

3) Nephrotoxicity: Metformin and Nateglinide were inactive for nephrotoxicity, whereas the other agents displayed active nephrotoxicity.

4) Respiratory Toxicity: Metformin was the only agent that showed inactive respiratory toxicity, while the other agents exhibited active respiratory toxicity.

5) Cardiotoxicity: Most of the antidiabetic agents were predicted to have inactive cardiotoxicity, with the exception of Acarbose, Voglibose, Dapagliflozin, Canagliflozin, Boxagliflozin, and Nateglinide, which were found to be cardiotoxic.

Based on the data presented in (Table 3), the majority of the selected antidiabetic agents exhibit organ-specific toxicities.

The toxicity end points of selected antidiabetic agents were predicted using the ProTox 3.0 web server, with a focus on Carcinogenicity, Immunotoxicity, Mutagenicity, Cytotoxicity and blood brain barrier. From the data presented in (Table 4) carcinogenicity is predicted to be active for Tenegliptin and Rasiglitazone, whereas the rest of the antidiabetic agents show inactive carcinogenicity. Immunotoxicity is observed in Acarbose and Voglibose, with the remaining agents displaying inactive immunotoxic potential. All selected antidiabetic agents demonstrate inactive mutagenicity and cytotoxicity. Additionally, Acarbose, Voglibose, and Repaglinide actively cross the blood-brain barrier, while the other agents do not exhibit this property. These predictions highlight specific toxicological and pharmacokinetic profiles among the selected antidiabetic agents.

CONCLUSION:

In this study, we have used computer-based methods to analyse the ADME of various antidiabetic drugs. We found that most agents had good absorption properties, but some showed potential toxic effects on specific organs. For example, Pioglitazone and Acarbose were linked to liver toxicity, while Sitagliptin and others could affect the nervous system. These findings highlight the importance of understanding how these drugs behave in the body and their potential side effects which mainly helps in data compilation about the selected antidiabetic agents which useful for the physician while prescribing the medications to diabetic patients.

FUTURE PERSPECTIVES:

Looking ahead, more extensive studies should be conducted to validate these predictions with experimental data. Researchers can further explore combining different computational techniques to improve accuracy in predicting drug safety and efficacy. Additionally, investigating new compounds through this in silico approach can help discover novel antidiabetic agents with better profiles and fewer side effects. Ultimately, this work can contribute to developing better treatments for diabetes that are both effective and safe for patients.

REFERENCES:

1. International Diabetes Federation. IDF Diabetes Atlas. 10th ed. 2021.

2. Saeedi P, Petersohn I, Salpea P, et al. Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: Results from the International Diabetes Federation Diabetes Atlas, 9th edition. Diabetes Res Clin Pract. 2019; 157: 107843.

3. Pansare K, Upasani C, Upaganlwar A, Sonawane G, Patil C. Pre-clinical study of lycopene alone and in combination with olive oil in streptozotocin-induced diabetic nephropathy. Vidyabharati Int Interdiscip Res J (Spec Issue). 2021 Aug; 320-332.

4. Bailey CJ. The current drug treatment landscape for diabetes and perspectives for the future. Clin Pharmacol Ther. 2017; 101(5): 535-547.

5. Rojas LBA, Gomes MB. Metformin: an old but still the best treatment for type 2 diabetes. Diabetol Metab Syndr. 2013; 5: 6.

6. Nathan DM, Buse JB, Davidson MB, et al. Management of hyperglycemia in type 2 diabetes: A consensus algorithm for the initiation and adjustment of therapy. Diabetes Care. 2009; 32(1): 193-203.

7. Nissen SE, Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med. 2007; 356(24): 2457-2471.

8. Lagorce D, Sperandio O, Baell JB, et al. FAF-Drugs3: A web server for compound property calculation and structure-based virtual screening. Nucleic Acids Res. 2015; 43(W1): W200-W207.

9. Daina A, Michielin O, Zoete V. SwissADME: A free web tool to evaluate pharmacokinetics, drug-likeness, and medicinal chemistry friendliness of small molecules. Sci Rep. 2017; 7(1): 42717.

10. Pires DEV, Blundell TL, Ascher DB. pkCSM: Predicting small-molecule pharmacokinetic and toxicity properties using graph-based signatures. J Med Chem. 2015; 58(9): 4066-4072.

11. Pathak S, Mishra A, Sonawane G, Sonawane K, Rawat S, Raizaday A, Singh SK, Gupta G. In silico pharmacology. In: Computational Approaches in Drug Discovery, Development and Systems Pharmacology. Academic Press. 2023; 1-52.

12. Sonawane G, Sharma S, Gilhotra R. In silico analysis of 1,3,4-oxadiazoles as potential BCL-2 inhibitor for cancer treatment. Asian J Chem. 2024; 36(5): 1072-1088.

13. Sonawane G, Sharma S, Gilhotra R. Synthesis, characterization and pharmacological screening of 1,3,4-oxadiazoles. Asian J Chem. 2024; 36(6): 1436-1446.

Received on 25.05.2025 Revised on 12.06.2025

Accepted on 29.06.2025 Published on 12.08.2025

Available online from August 18, 2025

Asian J. Research Chem.2025; 18(4):235-242.

DOI: 10.52711/0974-4150.2025.00036

This work is licensed under a Creative Commons Attribution-Non Commercial-Share Alike 4.0 International License. Creative Commons License.