IN SILICO AND IN VIVO SCREENING OF TRIAMTERENE SYNTHETIC ANALOGUES AS PROMISING DIURETICS

. In silico and in vivo screening of triamterene synthetic analogues as promising diuretics. Sokolova K.V., Kovalenko S.I. The modification of lead-compound aimed to the increasing of activity, decrement of toxicity or improvement of selectivity is one of the most important methods used for elaboration of novel medications. Natural compounds, approved or investigational drugs or just compounds with proved biological activity could be the lead-compound. Often the chemical modification of lead compounds is directed at the enhancement of ligand-biological target interactions. Abovementioned approach, namely structural modification of known drug triamterene was used for purposeful search for novel diuretics. The preliminary prognostication of ligand-target interactions and affinity levels allow to reduce quantity of experimental animals, synthesis, and pharmacological studies costs. Conducted studies revealed the series of promising 6,7-disubstituted pteridine-2,4(1H,3H)-diones with diuretic activity that comparable with pharmacological effect of triamterene. Aim – purposeful search for promising diuretics among structural analogues of triamterene that includes preliminary in silico studies, synthesis and in vivo screening of novel compounds for diuretic activity. organic synthesis, physicochemical methods of analysis of organic compounds (NMR 1 H-spectroscopy, chromato-mass spectrometry, elemental analysis). Prediction of affinity for a biological target, prediction of toxicity and lipophilicity of the combinatorial library, which was created on the basis of the drug triamterene, was carried out using computer services. Studies of compounds that affect the excretory function of the kidneys of rats were performed according to the generally accepted method of E.B. Berkhin with water load. Research of the probable mechanism was conducted by flexible molecular docking, as an approach of finding molecules with affinity to a specific biological target. Macromolecular data were downloaded from the Protein Data Bank (PDB) namely, the crystal structures of epithelial sodium channel (ENaC) ((PDB ID – 6WTH). The substantiation of potential diuretics design was conducted by in silico methods (prediction of affinity, ligand-enzyme interactions and pharmacokinetic characteristics). The structural modification of triamterene molecule was carried out by replacing of amino-group in positions 2, 4 and 7 by others “pharmacophore” fragments. Abovementioned transformation is aimed at the changing of ligand-enzyme interactions in active site, lipophility and toxicity. Synthesis of 6,7-disubstituted pteridine-2,4(1H,3H)-diones was conducted by condensation 5,6-diamino-2-oxo-(thioxo-)-2,3-dihydropyrimidin-4(1H)-ones with carbonyl-containing compounds or oxocarboxylic acids. The further modification of obtained compounds was performed by alkylation, hydrazinolysis and nucleophilic addition/elimination. The structure of obtained compounds was proven by elemental analysis, chromato-mass and 1 H NMR-spectral analysis. The studies of synthesized compounds effect on excretion function of kidneys allowed to detect series of promising structural analogues of triamterene that exceed it in pharmacological activity by 27.3-99.0%. The “structure-biological activity” relationship was discussed and perspective of the further search of diuretics among abovementioned compounds used to detect compounds with significant biological activity. Several effective compounds were identified among pteridines, which exceed the reference drug triamterene in terms of daily diuresis. The obtained results substantiate further purposeful search, in-depth research on experimental pathologies and study of the mechanism of action of potential diuretics among this class of compounds.

The process of new drugs developing has been significantly transformed today from in vivo models to "de novo design" methods, which are represented by directed design of ligands with high-affinity for biological targets involved in key stages of the pathogenesis of diseases [1]. The effectiveness of medications depends on their affinity for a protein or receptor, and molecules with low affinity will not be able to determine the required biological response. Thus, the affinity of a drug for a biological target is vital for predicting target-drug interactions, and allows a significant number of compounds to be studied prior to a traditional experiment.
One of the important biotargets is the epithelial sodium channels (ENaC), which are responsible for the reabsorption of sodium by the epithelium, lining the distal part of the renal tubules. In addition, ENaCs perform similar functional roles in some other tissues, such as the respiratory tract and the distal part of the colon [2, 3]. Sodium reabsorption is regulated by aldosterone, vasopressin and glucocorticoids and is one of the main mechanisms of regulation of sodium balance, blood volume and blood pressure. Sodium reabsorption is also inhibited by potassium-sparing diuretics: Amiloride and its analogues (Benzamil and Phenamil) and triamterene [4]. However, these drugs На умовах ліцензії CC BY 4.0 have side effects, namely hyperkalemia and urolithiasis, disorders of the gastrointestinal tract and central nervous system [5].
The choice of heterocyclic matrix, namely pteridine, which is present in the structure of known drug triamterene (6-phenylpteridine-2,4,7-triamine) was substantiated for the development of novel diuretics. Moreover, pteridines and related heterocycles are intensively studied in terms of the search for promising drugs [4,6]. Our previous studies have shown that for some 1-methyl-3-R-6-(2-hydroxy-(oxo-)-2-aryl-(hetaryl-)ethyl)pteridine-2,4,7 (1H, 3H, 8H)-triones is characterized by a high diuretic effect [7]. Therefore, the design and search for biologically active compounds with diuretic activity among the products of lead-compound triamterene modification is reasonable. The design of abovementioned compounds can be carried out by the replacement of amino groups of positions 2, 4 and 7 with other "pharmacophore" fragments in order to change the ligandenzyme interactions in the active site of the enzyme; introduction of other structural fragments to positions 2, 6 and 7 in order to change lipophilicity and toxicity (Fig. 1). This structural modification of the "leadcompound" can result in potentiation of the desired pharmacological activity, reduce toxicity and improve the selectivity of action. Therefore, the aim of present work is to continue our studies on the purposeful search for new biologically active compounds with diuretic activity among pteridines, namely the structural modification of the known drug triamterene using in silico methodologies and traditional synthesis and in vivo screening.

The general methods for the synthesis of 6-R 2 -7-R 3 -2-hydrazineyl-pteridin-4(3H)-ones (4.1-4.2).
To a suspension of 10 mmol of compound 3.1 or 3.4 in 20 ml of ethanol 1 ml (20 mmol) of hydrazine hydrate was added. The reaction mixture was refluxed for 1.5 h until the hydrogen sulfide evolution over. The reaction mixture was cooled, and the formed precipitate was filtered off and died.

2-((Furan-2-ylmethylene)hydrazineylidene)-6,7dimethyl-2,3-dihydropteridin-4(1H)-one
Ligand preparation. The substances were drawn using MarvinSketch 20.20.0 and saved in mol format [9]. After that they were optimized by program Chem3D, using molecular mechanical MM2 algorithm, and saved as pdb-files. Molecular mechanics was used to produce more realistic geometry values for most organic molecules, owing to the fact of being highly parameterized. Using AutoDockTools-1.5.6, the pdb-files were converted into PDBQT, number of active torsions was set as default [10].
Toxicity prognosis. Prediction of acute toxicity and calculation of octanol/water partition coefficient (logP) was made in silico using the service ProTox-II [12].
Study of the effect of compounds on the excretory function of the kidneys. The initial screening was performed on 174 white male Wistar rats weighing 120-170 g. The in-depth experiment was performed on 24 white male Wistar rats weighing 100-140 g, which 9 22/ Том XXVII / 3 were kept in standard conditions of the vivarium of the Dnipro State Medical University. Experimental studies were performed in accordance with the "General Ethical Principles of Animal Experiments" (Ukraine, 2001), the provisions of the "European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes" (Strasbourg, 1986) [13]. Screening of the new synthesized compounds, in order to identify diuretic properties in a few pteridine derivatives, was carried out according to the generally accepted method of E.B. Berkhin [14,15]. Prior to the experiment, the animals were kept without food for three hours. The diuretic effect of the compounds was studied under liquid load at the rate of 5 ml per 100 g of animal weight and without. The test compounds were administered to rats once intragastrically at a doses of 2.6 mg/kg body weight as an aqueous suspension simultaneously with the water load. Animals were placed in individual cages for urine collection during three hours and 24 hours. triamterene in equivalent doses for rats was selected as the reference drug [15].
The obtained data were statistically processed using the software package Statistica 6.1 (StatSoft Inc., serial number AGAR909E415822FA). The arithmetic mean values (M) and their errors (± m) were calculated. The probability of intergroup differences was determined using Student's parametric ttest and one-way analysis of variance (ANOVA). The differences were considered statistically significant at a value of p≤0.05 [16].

RESULTS AND DISCUSSION
According to the design of the present study the calculation of predicted affinity for biological target, predicted toxicity and lipophilicity were conducted for more then 50 compounds from combinatorial library based on triamterene structure. (Fig. 1). Molecular docking revealed that in most of cases structural modification of triamterene resulted the decreasing of studied compounds affinity for ENaC (

T a b l e 2
The main types of interactions of synthesized compounds and pharmacological standards with amino acid residues of epithelial sodium channel (ENaC)  The studied compounds have various interaction with amino acid moieties in active site of ENaC (Table 2). Hence, visualization of interaction of compound 2.1 with ENaC (Fig. 2, В) revealed the presence 22/ Том XXVII / 3 of one conventional hydrogen bond between NH-group in position with amino-acid moiety CYS267 (2,84Å). Besides, compound 2.1 forms three weaker carbon hydrogen bond between long pairs of N-5, O-4, NHgroup of pteridine cycle and SER210 (3,68Å), THR259 (3,52Å) and LEU257 (3,37Å), correspondingly. Additionally for compound 2.1 were predicted еlectrostatic (-anion) and hydrophobic (Alk) interactions between electron deficient cycle and ASP268 (3,78Å, 3,20Å), ALA269 (3,84Å) and LEU257 (3,37Å). Abovementioned interaction provide the placing of 2.1 molecule (LogP = -1.68) in lipophilic part of ENaC active site. It should be noted that placing of compound 2.1 is different from placing of triamterene molecule that despite higher lipophilicity (LogP = 2.58) located in more hydrophilic part of the ENaC pocket. Molecule of compound 2.4 (LogP =3.36) is placed in lipophilic part of active site of enzyme as well (Fig. 2, С). At the same time molecule of abovementioned compound form interaction with other amino acid moieties. (Table 2). Hence, molecular docking of compound 2.4 toward ENaC revealed (Fig. 2, С)  Molecule of compound 5.2 unlike compounds 2.1 and 2.5 is placed in hydrophilic part of ENaC active center (Fig. 2, D). Abovementioned placing is provided by two conventional hydrogen bonds of NHfragment in 1 st and 8 th position of the cycle and carbon hydrogen bond of carboxylic group with GLN334 (2,46Å and 2.41Å) and PRO338 (3,53 Å) correspondingly. Additionally placing of ligand in active site of enzyme is facilitated by -donor hydrogen bond and hydrophobic (-Alk) interaction of pteridine cycle with ASN288 (4,09Å and 3,54Å) and ALA387 (4,80Å), correspondingly.
In this way, conducted molecular docking of studied compounds toward biological target showed, that most of the compounds formed different from the triamterene ligand-enzyme interactions. At that time compound 5.2 has similar to triamterene position in active site of ENaC. Despite the obvious benefits, computer-aided drug design methods have some limitations. Hence, computer modelling methods do not consider diversity of the drug's effects on living organism of experimental animals. Traditional organic synthesis methods and their in vivo screening for biological activity should be conducted to estimate tha value of compounds as potential diuretics.

Fig. 3. Approaches to the synthesis of triamterene synthetic analogue's
The following step of chemical modification was the S-alkylation of compounds 2.8, 2.9, 2.10 and 2.12 with halogen-containing carboxylic acid. Abovementioned reaction was conducted according to standard procedure [20]. The 1 H NMR spectra of compounds 3.1-3.4, unlike spectra of compounds 2, were characterized by the presence of the signals that associated with substituents at sulfur atom, namely two proton singlet of SCH 2 -group at the 3.99-4,06 ppm. Besides, spectra of abovementioned compounds were characterized by the signals of exchangeable protons and set of the signals correspond to substituents at the positions 6 and 7 [19].

22/ Том XXVII / 3
It should be mentioned, that replacement of carboxyethylthiol fragment in position 2 of pteridine cycle of compound 3 by hydrazino-group proceeded easily. Abovementioned reaction required addition of 2-fold excess of hydrazine hydrate and short-term refluxing (Fig. 3). This reaction resulted in the formation of compounds 4 with satisfactory yields. Reaction of compounds 4 with furfural yielded corresponding hydrazones 5. Signals of hydrazinegroup protons were registered at the 7.40-7.22 ppm and 4.36-4.20 ppm in 1 H NMR spectra of compounds 4. At the same time spectra of compounds 5 were characterized by the singlet of azomethine proton at the 8.01-8.00 ppm. Presented data definitely prove the structure of obtained compounds.
The results of biological studies showed (Table 3), that most of the investigated compounds by the 2 hour of experiment revealed moderate diuretic effect or inhibited the diuresis of experimental animals that is characteristic for pteridine derivatives [4]. However, compounds 2.3 and 5.1 with diuretic effect comparable with activity of triamterene were found among the studied compounds. (Table 3). More interesting were the results of the study of synthesized compounds effect on day diuresis. It was found that compounds 2.2, 2.4, 2.7, 2.11 and 5.2 increased day diuresis by 76.4-148.1%, that exceeded activity of triamterene (49,1%, Table 3).

T a b l e 3
The effect of the synthesized compounds and reference drugs on the process of urination in intact rats under water load with a single injection (M±m, n=6)* Conducted QSAR studies showed that structural modification of triamterene via replacing of amino group in 2 nd and 4 th position by oxo-groups in most of cases was reasonable (compounds 2.1, 2.2 and 2.4).
Replacing of oxo-group in position 2 of pteridine system by thioxo-group (2.8-2.10), hydrazino-group На умовах ліцензії CC BY 4.0 (4. 1 and 4.2) led to the loss of diuretic activity. Transformation of thioxo-group at the 2 nd position into mercapthomethylcarboxylic fragment (compound 3.2) resulted in the decreasing of activity as well. Introduction of furylhydrazone fragment to position 2 of pteridine system (5.1 and 5.2) enhances diuresis. The nature of substituents in positions 6 and 7 have a significant impact on diuretic activity level.
Therefore, compounds 2.4 and 2.11 that contain arylvinyl fragments in positions 6 and 7 reveal significant diuretic activity. Hence, the purposeful search of biologically active compounds with diuretic activity among the structural analogues of triamterene is reasonable and opens prospects for further design, synthesis and advanced studies of their effect on excretion system.

CONCLUSIONS
The design of new biologically active compounds with diuretic activity was performed using in silico methodologies and realized by structural modification of the well-known diuretic triamterene. Condensation of 5,6-diamino-2-oxo-(thioxo-)-2,3dihydropyrimidin-4(1H)-ones with carbonyl containing compounds is convenient method for synthesis of promising diuretics that can be used for preparation of various by structure compounds. Several effective substances were identified among synthesized pteridines, which exceed the reference drug triamterene by the level of diuretic activity. Replacing of amino group in 2 nd and 4 th position by oxo-groups, introduction of furylhydrazone fragment to position 2 of pteridine system and arylvinyl fragments to positions 6 and 7 are reasonable in scope of novel diuretics synthesis while replacing oxo-group in position 2 of pteridine system by thioxo-group hydrazino-group, as well, as transformation of thioxo-group in the 2 nd position into mercapthomethylcarboxylic fragment lead to the loss of diuretic activity. The obtained results substantiate further purposeful search, in-depth research on experimental pathologies and study of the mechanism of action of potential diuretics among this class of compounds.

Recommendations.
The results of studies confirmed the diuretic effect of some 6-substituted pteridine-2,4,7 (1H, 3H, 8H) -triones and open up prospects for further study of their effects on the urinary system. In vitro study of synthesized compounds effect on molecular targets associated with diuretic activity, structural modification of active compounds by introducing of additional pharmacophore groups or their "bioisosteric" substitutions are among reasonable directions of further studies.