Covalent Docking for Novel Series of Spiro-butenolides on Trx R1 for QSAR Modelling

1. INTRODUCTION

Thioredoxin reductase I (Trx R1) is an enzyme of the redox homeostasis system [1, 2]. In the last two decades, medicinal chemists and pharmacologists have shown great interest in inhibitors of this enzyme as therapeutic agents [3-7]. A number of Trx R1 inhibitors act by the mechanism of covalently binding to the selenocysteine residue at position 498, A, or C of the chain [8]. A common marker of Trx R1 inhibitory activity against organic compounds is the presence of a chemotype represented by electron-deficient olefins (Michael acceptors) [9-11]. Previously, Trx R1 inhibitors containing an electron-deficient olefin group, namely peptidomimetics based on (E)-4-(4-fluorophenyl)-4-oxobut-2-enoic acid were presented in a study [12]. The development and search for new compounds, the mechanism of action of which is based on the inhibition of Trx R1, requires solving the problem of selective action on the selenocysteine residue, since in living organisms cysteine residues and free SH-groups are present in proteins in large amounts [13]. The selectivity of Michael acceptors is analyzed using model systems based on cysteine and selenocysteine [14]. The obtained data on inhibitory activity against Trx R1 [15] led us to think about the possibility of constructing a QSAR model of covalent binding of inhibitors to Sec498 Trx R1. In our study, we used Trx R1 as a drug target because this newly anti-coogenic target for antimetastasis therapy and very important strategy in medicinal chemistry. The aim of this study is to conduct virtual screening based on compu- tational chemistry algorithms for the targeted search for novel thioredoxin reductase inhibitors, as similar methods have been successfully applied previously [16].

2. MATERIALS AND METHODS

2.8. Colorimetric Detection of Cell Lysate TrxR Activity

For the TrxR assay, lysate from SH-SY5Y cells was used. A-548 cells were washed twice with phosphate buffer solution (PBS) and harvested from a flask with a scraper. Cells were homogenized in ice-cold 50 mM potassium phosphate buffer (pH 7.4), containing 1 mM EDTA, with Ultra Turrax T25 (Ikawerk, Janke and Kunkel Inc., Staufen, Germany). The homogenates were centrifuged at 10.000xg for 15 min at 4 °C. The protein concentrations of the supernatants were determined using a Bio-Rad protein assay kit (Bio-Rad Laboratories, USA). TrxR activity was assayed using 5,5- dithio-bis-(2-nitrobenzoic acid) (DTNB, Sigma-Aldrich, USA) as the substrate [12]. The reactions were run on 96 well plates, the final volume of 200 µl, in 50 mM potassium phosphate buffer (pH 7.4), containing 50 µg of the cell lysate proteins, 1mMEDTA, 50mMKCl, 0.2 mg/mL bovine serum albumin, 0.25 mM NADPH (Acros Organics, USA). The reaction mixture was incubated in 190 µl of reaction volume for 15 min on a plate shaker at room temperature, with compounds tested at appropriate concentrations. After incubation 10 µl of DTNB was added to a final concentration of 0.5 mM. Enzyme kinetics was monitored on the Clariostar microplate reader, by measuring the increase in absorbance at 412 nm for 20 min. The background TrxR-independent reduction of DTNB in the cell lysates was subtracted from each value. Aurum- tiomalate is used to validate inhibitor TRX.

Table 1.

S.No	Formula	IC50 µM
1		5.90 ± 0.69
2		1.49 ± 0.08
3		3.86 ± 0.25
4		4.92 ± 0.66
5		17.64 ± 0.74
6		53.71 ± 10.00
7		40.06 ± 6.00
8		8.87 ± 1.06
9		171.70 ± 14.00
10		117.64 ± 7.60
11		17.78 ± 1.54
12		26.07 ±3.04
13		37.15 ± 3.64
14		40.39± 2.35
15		40.45± 5.52
16		60.44 ± 6.00
17		40.45± 5.52
18		60.44 ± 6.00
19		6.76 ± 0.65
20		149.03 ±9.00

3. RESULTS AND DISCUSSION

In our recent work on the inhibition activity of human thioredoxin reductase I by a new series of spirocyclic compounds, obtained using an innovative diazo-chemistry approach, selective inhibitors with high activity were discovered that selectively bind the selenol group of selenocysteine [21]. Measured IC50 value for training set present in tables supplementary information.

The structures of the Trx R1 inhibitors used for modeling and their half-inhibition concentration values. Calculation of the Gibson binding energy of the enzyme and inhibitor allows one to calculate the stability constant of the complex. This value strongly depends on the HOMO energy of the inhibitor, since the transition of an electron from HOMO to LUMO leads to a strong polarization of the double bond, a change in its order and, as a consequence, the formation of a covalent bond with the selenide ion Sec-498, the rate of this process determines the effectiveness of inhibition. Thus, our QSAR model representing inhibitor efficiency in the form of 1/lnIC50 values takes into account the contribution of the following molecular descriptors: Muliken charge, the electrostatic potential difference on beta and alpha atoms, Ovality, Electrostatic potential and HOMO energy [22, 23].

Table 2 presents the results of binding energy for covalent docking with a molecular descriptor.

The regression of the predicted half-inhibition concentration values with the experimental values and the pharmacophore model are presented in Fig. (1).

CONCLUSION

As a result of modeling the binding process of electron-deficient olefins in the active center of thioredoxin reductase with a selenocysteine residue, we found that the rate of the binding process is limited by the formation of a complex between the inhibitor and the enzyme. Interaction with residues Lys-29, Thr-481 and Gln – 494 ensures the necessary conformational changes of the inhibitor in the active site, which allows obtaining the necessary gain in energy to provide the binding of the inhibitor. The presence of a methoxy substituent in the aryl fragment near the reaction center is preferred. This fact is associated with ensuring optimal polarization of the double bond of the electron-deficient olefin. The sites of contact with the residues of the side groups of the active center must be aromatic and hydrophobic, at the same time, the conformational mobility of the double bond of the electron-deficient olefin provides sufficient freedom for successful interaction with the selenium atom. The presence of donor substituents at the nitrogen atom of the homophthalemide fragment leads to a decrease in the electrostatic charge on the beta atom of furanone, which reduces its electrophilicity and, as a consequence, the rate of reaction with the selenium atom SEC 498. Alkyl substituents are less preferred in this case. In general, QSAR analysis allows us to present a pharmacophore model for unsaturated spiranolactones and furanones. At the moment, our group has obtained an extensive library of new spirobutenolides, as well as new electron-deficient olefins based on benzylidene succinimides, we plan for 120 new compounds to calculate molecular descriptors and semi-inhibition concentration values using our model, perform a colorimetric determination of inhibition activity in a wet experiment and test the stability and the accuracy of the model, which we hope to publish in the next article. This publication opens a series of works on the rational design of thioredoxin reductase inhibitors using methods QSAR.

AUTHOR’S CONTRIBUTION

It is hereby acknowledged that all authors have accepted responsibility for the manuscript's content and consented to its submission. They have meticulously reviewed all results and unanimously approved the final version of the manuscript.

REFERENCES