Repurposing of Fluindione an Anti-Coagulant Drug Against Thymondylate Synthetase For Skin Cancer Management

Sonakshi Lokare, Sakshi Khamkar, Sankalp Karande, Sanika Mithari, Sayali kawade, Namrata koli, Ankita Kharage , Lohar Dayanand, Kukade Akanksha,

doi:10.5281/zenodo.15315245

Research Paper | Open Access
Volume 02 | Issue 05 | Article Id IJSRT/250304201

Repurposing of Fluindione an Anti-Coagulant Drug Against Thymondylate Synthetase For Skin Cancer Management
Sonakshi Lokare* Sakshi Khamkar Sankalp Karande Sanika Mithari Sayali kawade Namrata koli Ankita Kharage Lohar Dayanand Kukade Akanksha
Ashokrao Mane College of Pharmacy, Peth Vadgaon, India

Abstract

The drug repurposing of previously approved drugs is an attractive way to facilitate the development of new drugs for treating cancer. This study examined the feasibility of using the existing anticoagulant drug Fluindione as a potential inhibitor of thymidylate synthase (TS) for the treatment of skin cancer. Ligand-based virtual screening drug repurposing and molecular docking studies have revealed that Fluindione (DrugBank ID: DB13136) has a high binding affinity for TS and ranks as the leading compound for docking with a score of -7.6. The structure docked suggested Fluindione binds with C3 of the TS enzyme, a large binding pocket that is well-defined and accessible. This observation supports the potential for Fluindione to be an effective inhibitor of TS. Stabilization characteristics apparent from the docking characteristics of Fluindione and comparison features to the therapeutically relevant compound Fluorouracil indicate that fluindione may offer a more advantageous treatment opportunity as a single agent, or in conjunction with standard therapy in skin cancer. Overall, these findings provide strong evidence for the implementation of drug repossession drug design studies and suggest impetus for confirming Fluindione's anticancer potential through further research.

Keywords

Drug repossession, Fluindione, thymidylate synthase, skin cancer, molecular docking, anticoagulant, virtual screening, cancer therapeutics

Introduction

Fluindione, a vitamin K antagonist which has traditionally been used for its anticoagulant properties, is clinically deemed useful for treatment of thromboembolic diseases due to its effect upon inhibiting vitamin K epoxide reductase and other enzymes responsible for the production of essential platelet reliant clotting factors [1]. Although fluindione is effective as an anticoagulant, clinical application is limited by its narrow therapeutic range and high sensitivity to drug and dietary binding [2]. Recently, repurposing fluindione for oncological indications has arisen given evidence that fluindione may exert non-anticoagulant effects [3]. This is especially relevant when considering skin cancers, the most qualified cancer globally, with NMSC including basal cell carcinoma (BCC) and squamous cell carcinoma (SCC), and melanoma being the most challenging skin cancer [4]. While surgically removable early-detected NMSCs are considered treatable, advanced melanoma poses a significant treatment hurdle due to its high metastatic potential and de nova resistance to standard chemotherapy regimens and immunotherapy [5]. With this in mind, TS (thymidylate synthase) has become an attractive target as an enzyme important in the synthesis of DNA, expressed in excess by rapidly proliferating cancer cells [6]. Inhibition of TS interrupts DNA replication, thereby forcing cell cycle arrest or apoptosis. The present TS inhibitors, 5-fluorouracil, for example, show the principle, but we still have issues of efficacy and toxicity [7]. Fluindione has the potential to inhibit TS, and therefore provides a new option for skin cancers, especially when current treatments are inadequate [8]. Drug repurposing can provide a cost and time advantage to research in cancer in general, and fluindione has an existing safety profile that makes it a good candidate [9]. In this review, we examine the new evidence suggesting repurposing fluindione in skin cancer therapy focusing on the mechanisms of actions, particularly TS inhibition, and the designed repair of improved treatment outcomes in melanoma and skin cancer [10].

MATERIALS AND METHODS:

Data Collection: Sources of Fluindione Analogs

We started with compounds from a few reliable databases: PubChem, very large and comprehensive, harboring all chemical information available, including a wide array of derivatives [11], ChEMBL for molecules with drug-like properties and established biological activities, and the ZINC Database, a free source of commercially available compounds that provides a large collection of Fluindione analogs [12]. Besides, we carried out a review of published articles and patents to collect even more derivatives of Fluindione, other than those mentioned in the above databases [13].

Ligand – Based virtual screening protocol.

In ligand-based virtual screening, compounds were selected on multiple criteria to highlight potential drug candidates[14]. Chemical diversity was ensured through the assessment of structural variety with Tanimoto similarity coefficients ADMET properties were assessed to rank compounds according to favorable pharmacokinetic profiles for oral bioavailability[15]; biological relevance was taken into account by considering compounds with reported activity against the target of interest[16]; molecular weight between 300-500 Da and lipophilicity within LogP values 2-5 as optimized for pharmacological properties, and availability that allowed ease of experimental validation[17]. Candidates for drug repurposing were tested on Ligand-based Screen by using ligand-based virtual screening [18]. The approach made use of a known active ligand of an exemplified target as a template compound, and it was used as a template for screening by different similarity measureslike LigMate, FitDock-align, Morgan Fingerprint, LSalign, FP2, and FP4, to rank compounds based on their similarity scores and then to pick the top-ranked compounds for further studies in drug repurposing [19].

Docking studies

In one previous drug repurposing project, I attempted to generate a compound library from databases like ZINC and ChEMBL [20]. Subsequently, I constructed a pharmacophore model using the active ligand in order to carry out virtual screening using molecular docking and pharmacophore-based methods [21]. After post-docking analysis was done for evaluation of binding modes and interactions, I prioritized high-scoring compounds further for evaluation [22].

RESULTS AND DISCUSSION

Results of ligand-based drug repurposing

Table 1: justification for selecting fluorouracil as the primary ligand for designing new agents targeting skin cancer.

Criteria	Justification
Mechanism of Action	5-FU is a well-established chemotherapeutic agent that inhibits thymidylate synthase, impairing DNA synthesis and cell proliferation, which is effective in cancer treatment.
Target Specificity	Fluorouracil has been shown to specifically target cancer cells, including skin cancer, by interfering with nucleic acid metabolism.
Efficacy in Skin Cancer	5-FU has demonstrated efficacy in treating non-melanoma skin cancers, such as basal cell carcinoma and squamous cell carcinoma.
Clinical Experience	Fluorouracil is FDA-approved for topical and systemic use in skin cancer treatment, indicating its safety and effectiveness.
Synergistic Potential	5-FU can potentially be combined with other therapies (radiation, immune checkpoint inhibitors) to enhance the treatment efficacy for skin cancer.
Selectivity and Safety Profile	Topical 5-FU treatment minimizes systemic side effects, making it safer for localized skin applications.
Formulation Flexibility	Available in various forms (cream, injection), allowing for adaptable delivery methods to target skin cancer.
Resistance Consideration	While resistance can develop, modifications to 5-FU or its delivery can overcome such challenges.

Table (1) gives the justification for selecting fluorouracil as the primary ligand for designing new agents targeting skin cancer.

Result of ligand-based screening using the drug rep platform

Table 2: matching scores and target interactions of various compounds

Rank	Compound	Name	Score	Rank	Compound	Name	Score
1	DB00544	Fluorouracil	0.263	11	DB00170	Menadione	0.154
2	DB12466	Favipiravir	0.205	12	DB03209	Oteracil	0.152
3	DB01099	Flucytosine	0.200	13	DB00201	Caffeine	0.151
4	DB09256	Tegafur	0.189	14	DB00824	Enprofylline	0.145
5	DB09327	Tegafur-uracil	0.189	15	DB00832	Phensuximide	0.145
6	DB00322	Floxuridine	0.167	16	DB15598	Ferric maltol	0.143
7	DB01223	Aminophylline	0.157	17	DB09257	Gimeracil	0.143
8	DB00277	Theophylline	0.157	18	DB05246	Methsuximide	0.140
9	DB13228	Flosequinan	0.155	19	DB00791	Uracil mustard	0.140
10	DB13136	Fluindione	0.154	20	DB00347	Trimethadione	0.140

The table (2) The data indicates that the highest scoring compound in this dataset is DB00544 (Fluorouracil) with a score of 0.263, this compound has the highest score in this dataset. This means that Fluorouracil has the highest binding affinity or binding interaction according to the scoring system used in this thesis. Second place is DB12466 (Favipiravir) with a score of 0.205 and third place is DB01 099 (Flucytosine) with a score of 0.200. The lowest scoring compound in this dataset is DB00347 (Trimethadione) with a score of 0.140. Overall, the compounds vary by binding strength, while Fluorouracil clearly stood out as the highest scoring compound in table. This means Fluorouracil likely had the most favorable interaction in whatever context is being examined.

Table 3: Validation metrics from PDB-REDO

Validation Metrics	Original	PDB-REDO
Crystallographic refinement
R	0.2197	0.1702
R-free	0.2621	0.2233
Bond length RMS Z-Score	0.639	0.207
Bond angle RMS Z-Score	0.867	0.434
Model quality raw scorespercentiles
Ramachandran plot normality	6	38
Rotamer normality	16	64
Coarse packing	8	12
Fine packing	45	81
Bump severity	43	43
Hydrogen bond satisfaction	15	25

Comparisons of crystallographic refinement information supports relief in the overall structures for both PDB-REDO models. The R factor shows that the original model was originally 0.2197, and the PDB-REDO model is 0.1702 (which is a better fit to the model to the observed data). The R-free value also decreased from 0.2621 to 0.2233 compared to the PDB-REDO model which will also indicate better refinement and validation of the model. The Bond length RMS Z-Score also shows a substantial improvement from 0.639 to 0.207 which indicates better accuracy in the rectilinear representation of the bond lengths. Similar improvements can be seen from the Bond angle RMS Z-Score from 0.867 to 0.434 which suggests after optimally refining the PDB-REDO model we see that the geometry slightly improved in the depiction/representation of well parameterized values. In regards to model quality, the Ramachandran plot normality was also improved as, PDB-REDO model 38 vs original model of 6 (backbone dihedral angles reasonably distributed). Rotamer normality improved from 16 to 64 which is also good to see that the side-chain conformations appear better in the overall model. Additionally, fine packing did improve notable from 45 to 81 which suggests we have44 improved the overall efficiency of packing of structural support but still have areas of opportunity for the structural geometry.

Figure 1: Comparative Analysis of Model Quality Metrics: Original vs. PDB-REDO Refinement

Figure 2: Kleywegt- like plot

Figure (2) presents a kleywegt-like plot depicting the distribution of phi (Φ) and psi (Ψ) dihedral angles of amino acids in a protein, where the color gradient represents a density distribution of conformational states. The red areas indicate the favored conformations, while the scattered blue spots and orange dots represent individual residues I observed for stereochemical quality and structural integrity of the protein backbone.

Docking Results:

Table 4: Binding Affinity Analysis of Drugbank compounds to target pockets: Identification of potential drug candidates.

Drug bank ID	Pocket	Score	Chain	Drug bank ID	Pocket	Score	Chain
DB00544	C5	-4.5	A	DB00170	C3	-6.3	A
DB12466	C5	-5.3	A	DB03209	C5	-5.4	A
DB01099	C4	-4.7	A	DB00201	C3	-5.3	A
DB09256	C3	-5.7	A	DB00824	C4	-5.7	A
DB09327	C2	-5.5	A	DB00832	C3	-6.0	A
DB00322	C4	-6.3	A	DB15598	C3	-4.5	A
DB01223	C3	-5.3	A	DB09257	C4	-4.4	A
DB00277	C3	-5.3	A	DB05246	C3	-7.0	A
DB13228	C3	-6.7	A	DB00791	C4	-5.4	A
DB13136	C3	-7.6	A	DB00347	C4	-4.6	A

The table (4) shows that the drug that interacts most favorably, the lowest score, is DB13136 in Chain C3 with a score of -7.6. In fact, this is the most favorable interaction in overall trickless interactions, even with the negative score. Most of the drugs are associated with Chain A, which suggests that drugs have a consistent binding preference for Chain A. Though the drug DB05246 with a score of -7.0 also showed an interaction that was relatively strong, the score was lower than DB13136. Conversely, the drug DB15598 with a score of -4.5 was weaker overall. In conclusion, while the data shows that all of the drugs had negative scores, DB13136 provided the highest noted binding affinity for the listed candidates

Figure 3: Cavities found in protein

The data (figure 3) indicates a fluctuation of cavity volumes for different pocket IDs. C1 is the largest cavity volume with 547 Å³ and could hold greater molecules or interactions than the others. C5 is the smallest volume with 126 Å³ and likely would have the least space for molecular binding. C2 and C3 offered intermediate cavity volumes of 274 Å³ and 257 Å³, respectively, indicating a descriptive capability for binding not unlike that of C1. C4 was not quite as small with a cavity volume of 221 Å³ and would be less; however, it certainly would have some molecular interactions. The general cavity volume differences may lead to different interactions with these different pockets. The biggest cavity by volume is C1, and the smallest cavity is C5.

Figure 4: Molecular Docking of Fluindione: Key Binding Interaction for Skin Cancer

CONCLUSION

Fluorouracil (5-FU) is a great lead candidate for development in skin cancers because of understanding its mechanism of action as thymidylate synthase inhibitor, which disrupts DNA synthesis thereby reducing a cell's ability to proliferate. Its clinical use is against non-melanoma skin cancers and it is an accepted agent, having received FDA approval for topical and systemic use, so safe use in humans would be inferred. As an emerging strong binder, fluorouracil scored significantly from the ligand-based screening representing, stronger than any other potential candidates and it had the highest ligand-based scoring interaction score of 0.263. The use of PDB-REDO refinement was also helpful and a significant amount of the quality was restored to the protein model for docking. The PDB-REDO refined model R-factors stated improved R-factors and reduced R-factors as well as better geometric parameters lending confidence to the docking scores. In spite of 5-FU's proven clinical history, there remained some underappreciated possibilities that had greater docking interfaces, including DB13136 (Fluindione; -7.6) and DB05246 (Methsuximide; -7.0). In fact, both of these alternatives also have complimentarily larger protein pockets (C3), whereas 5-FU had an interface score of -4.5 with the smallest pocket (C5). The difference in scoring may contribute to the strength of the bonding of the compound. Despite being less favoured with regards to the docking score, the same clinical history that provided safer and other potential formulations to Fluorouracil, keeps it as an eligible lead compound for drug development. We recommend some alternative candidates, DB13136 (Fluindione) and DB05246 (Methsuximide) be developed as either a complementary active agent or as potential alternative active agents.

REFERENCE

Bossavy JP, Sakariassen KS, Thalamas C, Boneu B, Cadroy Y. Antithrombotic efficacy of the vitamin K antagonist fluindione in a human ex vivo model of arterial thrombosis: effect of anticoagulation level and combination therapy with aspirin. Arteriosclerosis, thrombosis, and vascular biology. 1999 Sep;19(9):2269-75.
Cazaux V, Gauthier B, Elias A, Lefebvre D, Tredez J, Nguyen F, Cambus JP, Boneu B, Boccalon H. Predicting daily maintenance dose of fluindione, an oral anticoagulant drug. Thrombosis and haemostasis. 1996;75(05):731-3.
Comets E, Diquet B, Legrain S, Huisse MG, Godon A, Bruhat C, Chauveheid MP, Delpierre S, Duval X, Berrut G, Verstuyft C. Pharmacokinetic and pharmacodynamic variability of fluindione in octogenarians. Clinical Pharmacology & Therapeutics. 2012 May;91(5):777-86.
Fahradyan A, Howell AC, Wolfswinkel EM, Tsuha M, Sheth P, Wong AK. Updates on the management of non-melanoma skin cancer (NMSC). InHealthcare 2017 Nov 1 (Vol. 5, No. 4, p. 82). MDPI.
Salas-Benito D, Pérez-Gracia JL, Ponz-Sarvisé M, Rodriguez-Ruiz ME, Martínez-Forero I, Castañón E, López-Picazo JM, Sanmamed MF, Melero I. Paradigms on immunotherapy combinations with chemotherapy. Cancer discovery. 2021 Jun 1;11(6):1353-67.
Rahman L, Voeller D, Rahman M, Lipkowitz S, Allegra C, Barrett JC, Kaye FJ, Zajac-Kaye M. Thymidylate synthase as an oncogene: a novel role for an essential DNA synthesis enzyme. Cancer cell. 2004 Apr 1;5(4):341-51.
Van Triest B, Pinedo HM, Giaccone G, Peters GJ. Downstream molecular determinants of response to 5-fluorouracil and antifolate thymidylate synthase inhibitors. Annals of Oncology. 2000 Apr 1;11(4):385-91.
Nair AS, Singh AK, Kumar A, Kumar S, Sukumaran S, Koyiparambath VP, Pappachen LK, Rangarajan TM, Kim H, Mathew B. FDA-approved trifluoromethyl group-containing drugs: A review of 20 years. Processes. 2022 Oct 11;10(10):2054.
De Lellis L, Veschi S, Tinari N, Mokini Z, Carradori S, Brocco D, Florio R, Grassadonia A, Cama A. Drug repurposing, an attractive strategy in pancreatic cancer treatment: preclinical and clinical updates. Cancers. 2021 Aug 5;13(16):3946.
Kumbhar P, Kole K, Yadav T, Bhavar A, Waghmare P, Bhokare R, Manjappa A, Jha NK, Chellappan DK, Shinde S, Singh SK. Drug repurposing: An emerging strategy in alleviating skin cancer. European Journal of Pharmacology. 2022 Jul 5; 926:175031.
Williams AJ. Public chemical compound databases. Current Opinion in Drug Discovery and Development. 2008 May 1;11(3):393.
Sterling T, Irwin JJ. ZINC 15–ligand discovery for everyone. Journal of chemical information and modeling. 2015 Nov 23;55(11):2324-37.
Zhou Y, Wang J, Gu Z, Wang S, Zhu W, Acen?a JL, Soloshonok VA, Izawa K, Liu H. Next generation of fluorine-containing pharmaceuticals, compounds currently in phase II–III clinical trials of major pharmaceutical companies: new structural trends and therapeutic areas. Chemical reviews. 2016 Jan 27;116(2):422-518.
Bhunia SS, Saxena M, Saxena AK. Ligand-and structure-based virtual screening in drug discovery. InBiophysical and Computational Tools in Drug Discovery 2021 Aug 7 (pp. 281-339). Cham: Springer International Publishing.
Maliehe TS, Tsilo PH, Shandu JS. Computational evaluation of ADMET properties and bioactive score of compounds from Encephalartos ferox. Pharmacognosy Journal. 2020;12(6).
Petrone PM, Simms B, Nigsch F, Lounkine E, Kutchukian P, Cornett A, Deng Z, Davies JW, Jenkins JL, Glick M. Rethinking molecular similarity: comparing compounds on the basis of biological activity. ACS chemical biology. 2012 Aug 17;7(8):1399-409.
Waring MJ. Defining optimum lipophilicity and molecular weight ranges for drug candidates—molecular weight dependent lower log D limits based on permeability. Bioorganic & medicinal chemistry letters. 2009 May 15;19(10):2844-51.
Ferraz WR, Gomes RA, S Novaes AL, Goulart Trossini GH. Ligand and structure-based virtual screening applied to the SARS-CoV-2 main protease: an in-silico repurposing study. Future medicinal chemistry. 2020 Oct 1;12(20):1815-28.
Srinivasarao M, Low PS. Ligand-targeted drug delivery. Chemical reviews. 2017 Oct 11;117(19):12133-64.
Martorana A, Perricone U, Lauria A. The repurposing of old drugs or unsuccessful lead compounds by in silico approaches: new advances and perspectives. Current Topics in Medicinal Chemistry. 2016 Aug 1;16(19):2088-106.
Pal S, Kumar V, Kundu B, Bhattacharya D, Preethy N, Reddy MP, Talukdar A. Ligand-based pharmacophore modeling, virtual screening and molecular docking studies for discovery of potential topoisomerase I inhibitors. Computational and structural biotechnology journal. 2019 Jan 1;17:291-310.
Blanes-Mira C, Fernández-Aguado P, de Andrés-López J, Fernández-Carvajal A, Ferrer-Montiel A, Fernández-Ballester G. Comprehensive survey of consensus docking for high-throughput virtual screening. Molecules. 2022 Dec 25;28(1):175.