Searching for novel antagonists of adenosine A1 receptors among azolo[1,5-a]pyrimidine nitro derivatives

Introduction: Ligands of adenosine A1Rs are potential candidates for the development of drugs for the treatment of paroxysmal supraventricular tachycardia, angina pectoris, hypertriglyceridemia, type 2 diabetes mellitus, neuropathic pain, and heart failure. At the same time, there is a deficiency of drugs that can regulate the functions of A1 receptors. A number of A1-antagonists are at the various stages of clinical trials; other drugs are not very selective or are characterized by an insufficient breadth of their therapeutic action. Therefore, the search for new medicinal compounds for the prevention and treatment of A1-depended diseases among nitro derivatives of tetrazolo[1,5-a]pyrimidine and 1,2,4-triazolo[1,5-a]pyrimidine is of scientific interest. Materials and methods: The search for active compounds was carried out by in silico and in vitro methods. At the first stage, a computer forecast of A1-antagonistic activity was carried out using the Microcosm BioS software. At the second stage, the prediction results were verified in vitro in a model of isolated mouse atria. Results and discussion: Based on the results of the prediction by the method of maximum similarity to standards, the most active compounds III, VIII, and XVII were selected. After testing the prediction results by the isolated atria method, the compound VIII was characterized by A1-blocking effect in vitro at a concentration of 10 μmol/L. Conclusion: The most promising compound with A1-blocking effect in vitro was identified; it is a derivative of tetrazolo[1,5-a]pyrimidine under the code of VIII. It is of interest for us for further in-depth study of its pharmacological properties.


Introduction
Adenosine receptors (ARs) are targets for the development of new methods of pharmacological correction of various diseases (Manjunath et al. 2009;Jamwal et al. 2019). Adenosine is an endogenous purinebased nucleoside expressed nearly in all body tissues (Saini et al. 2022). It exhibits multiple biological effects and regulates a number of physiological functions, including the functioning of the cardiovascular system (Headrick et al. 2011;Albrecht-Küpper et al. 2012;Guieu et al. 2020). It is known that adenosine regulates myocardial function and the intensity of coronary circulation, and also has a pronounced vasodilatory effect (Reiss et al. 2019). At the same time, the clinical potential of A 1 Rs regulation has not been fully realized to date. In modern medicine, only a few drugs with A 1 antagonistic properties are used.
Caffeine causes most of its biological effects via antagonizing all types of ARs. When acting as nonselective AR antagonist (Chen et al. 2017), caffeine, used acutely, is doing the opposite of activation of adenosine receptors, due to removal of the adenosinergic tonus (Ribeiro and Sebastião 2010). Synthetic xanthines, analogues of caffeine -Tonapofylline (BG-9928), Rolofylline  and Derenofylline (SLV-320) were first announced in 2010-2012 and went through the stages of clinical trials (Sureechatchaiyan et al. 2018) for the treatment of congestive heart failure, acute decompensated heart failure and renal dysfunction (Pang et al. 2011). These compounds are more selective A 1 R antagonists (Ensor and Russell 2010;Teerlink et al. 2012).
Single administration of Rolofylline in patients with heart failure and impaired renal function produced a consistently increased diuresis and natriuresis without compromising renal function (Ponikowski et al. 2010). In phase III clinical trials, the A 1 R antagonist Rolofylline was not sufficiently effective in worsening renal function in patients with acute heart failure (Voors et al. 2011). Clinical trials of Derenophylline and Tonapofylline were terminated for reasons beyond the effects of the drugs. Taking this into account, the issue of developing and creating new A 1 -antagonists remains very relevant.
In this regard, this study is focused on continuing the search for new A 1 R antagonists among pyrimidine derivatives as possible drug candidates for subsequent use in medical practice. Analysis of the chemical structure of known A 1 -antagonists (Tonapofylline and Rolofylline are derivatives of imidazopyrimidines and Derenofylline is pyrrolo[2,3-d]pyrimidine) suggests a high prospect of searching for potential A 1 R antagonists among pyrimidine derivatives. The current investigation is aimed to study the A 1 R antagonism potential of earlier synthesized nitro derivatives of tetrazolo[1,5-a]pyrimidine (Savateev et al. 2018) and 1,2,4-triazolo[1,5-a]pyrimidine (Rusinov et al. 2017).

Experimental design
At the first stage, a database was formed in silico on the structure of the reference ligands of adenosine A 1 receptors studied by the world scientific community using the search engines IUPHAR, Tocris, SelleckChem. Thus, 20 selective A 1 adenosine receptor antagonists were selected. Then, a database was formed on the structure of 25 new chemical compounds, derivatives of azolo[1,5-a]pyrimidine. The data on the structure of 25 new compounds were collected and generated using ChemFinder 9.0, a ChemOffice 9.0 software package (URL: http://www. cambridgesoft.com/12).
The processing of the data set and their translation into QL language descriptors were carried out using the utilities of the IT Microcosm software package; the following modules were used: ActUtil, TranQL2, and MakeData (Vassiliev et al. 2014). First, using the ActUtil utility, a data set consisting of * .actv and * .tba files was created. Second, the VMNC.sdf file was processed in the TranQL2 module to obtain * .qll files. At the last stage, the final data set was obtained using the MakeData utility.
The most active substances according to in silico forecast were tested for their A 1 -antagonistic activity with isolated tissue technique in vitro.

Predicted affinity for adenosine A 1 receptors of new compounds using the Microcosm BioS software
In this study, the prediction of affinity for adenosine A 1 receptors was carried out using the Microcosm BioS v.18.1.9 system, which includes the original QSAR database containing verified, structured and processed information on the chemical structure and activity level of the known compounds studied by the world scientific community at various types of biological activity.
The specified QSAR database was formed outside the scope of this study, using information obtained from three well-known international search engines: ChEM-BL (URL: https://www.ebi.ac.uk/chembl/), BindingDB (URL: https://www.bindingdb.org/) and PubChem (URL: https://pubchem.ncbi.nlm.nih.gov). The primary dataset was checked for correct chemical structural formulas, so incorrect and duplicate entries were removed from it.

Experimental animals
The experiments were carried out on 30 white outbred mice m=18-30 g, obtained from the Rappolovo Nursery, Leningrad region (Russia). The animals were kept in a vivarium with a natural light regimen on a standard diet of laboratory animals, with access to food and water ad libitum. The vivarium maintained standard conditions in accordance with Order No. 51 of August 29, 2014 of the Chief State Sanitary Doctor of the Russian Federation "On Approval of Sanitary Regulations 2.2.1.3218-14 "Sanitary and Epidemiological Requirements for Managing, Equipping and Maintaining Experimental Biological Clinics (Vivariums)" on a special hygienic wood filler (LLC Production Complex GlavRezerv). Twelve hours before the start of the experiment, the animals were deprived of access to food, while access to water remained free. The experimental procedures on animals were carried out in accordance with the Local Ethics Committee of Volgograd State Medical University, Volgograd, Russia (Minutes No. IRB 00005839 IORG 0004900 (OHRP)). All the experiments were carried out in accordance with "The Guidelines for Conducting Preclinical Studies of Drugs" (Mironov 2012).

Isolated atria technique
After cervical dislocation of the rodents, the heart was removed and both atria were isolated. The isolated samples of the left and right atria were fixed in a 20 ml bath, filled with Krebs-Henseleit nutrient solution with constant oxygenation of 95% O 2 -5% CO 2 and a temperature of 37 °C in a 4-channel system for maintaining the vital functions of isolated tissues TISSUEBATH4 (Biopac Systems, Inc., USA). The isolated atria were fixed on an isometric sensor TSD125C with a sensitivity range of 0-50 g at an isometric load of 1 g. The organ was left for adaptation for 45-60 min before the start of the experiment. During the adaptation period and the subsequent experiment, the buffer solution in the bath was replaced every 10-15 min.

A 1 -antagonistic activity of compounds under study
The A 1 -antagonistic activity of novel nitro derivatives of azolo[1,5-a]pyrimidines was investigated by addition of the tested compound (200 μl) 3 min before the introduction of adenosine (200 μl) with an exposure for 30 s before a registered decrease in chronotropism of atria, working in their own rhythm (without stimulation). Between each administration of adenosine, the isolated atrial sample was left to relax for at least 10-15 min to return the atrial chronotropism to the initial level. The reference drug was caffeine at an equimolar concentration of 10 μmol/L. The A 1 -antagonistic effect of the test substances and the control sample was assessed by a degree of suppression of adenosine-induced decrease in chronotropism of the isolated atrial tissue in comparison with the effect of adenosine obtained in control measurements (Δ%).
Contractions of isolated atria were recorded using Ac-qKnowledge 4.0 software (Biopac Systems, Inc., USA). The number of isolated atria contractions was measured over a 30-second interval with subsequent calculation of the heart rate. The amount of atrial chronotropism suppression (Δ%) was calculated by the formula: where ΔHR control -the difference between the baseline heart rate and the heart rate calculated after adenosine administration; ΔHR treated -the difference between the baseline heart rate and the heart rate calculated after adenosine administration in mice under the influence of the test compounds.

Statistical analysis
Statistical analysis of the obtained data was carried out in the GraphPad Prism 6.0 software using the Kruskal-Wallis test and post hoc Dunn's test. The data are presented as mean ± standard error of mean (M±SEM). The p<0.05 values were considered statistically significant.

Results of in silico prediction of the adenosine A 1 activity levels of new chemical compounds.
On this basis, the QSAR database was calculated using six specially created programs, containing the QL-descriptor representation of chemical structural information, in the form of working files in the format developed earlier when creating the IT Microcosm package (Vassiliev et al. 2014).
The QSAR database obtained in this way includes information on the structure of 625888 known chemical compounds tested by world scientific studies for 11509 different types of biological activity, with details of a type of biotarget, test organism and the measured activity indicator. In total, this database contains information on 6016 biotargets, on 325 species of test organisms, indicating the quantitative values of activity for 4 different indicators.
The specified QSAR base is part of the Microcosm BioS system, designed to predict the spectrum of targeted biological activity of new chemical compounds.
The QSAR-base of the Microcosm Bios v18.1.9 system contains information on the structure and level of affinity for the adenosine A 1 receptor of 7455 compounds studied in experimental tests on five different organisms, including data for 5373 compounds for Homo sapiens. The calculation of the predicted spectrum of targeted biological activity of new chemical compounds was carried out using the original Microcosm BioS system by the method of maximum similarity to standards. The well-known compounds studied by the world scientific community for A 1 -antagonistic activity were used as standards. Information about these compounds was accumulated in the above-described Microcosm BioS QSAR database.
The forecast results were saved as a tab-delimited text file; part of the primary listing is shown in Fig. 1.
The final results of the prediction in Microcosm BioS of new compounds adenosine A 1 receptor affinity are shown in Table 1.
In total, according to the prediction data by the method of maximum structural similarity to the standards, 7 compounds (highlighted in color) are promising for studying their adenosine A 1 receptor affinity, having three non-negative averaged estimates in total. The most interesting for experimental testing are 3 compounds (highlighted in green), in which all three calculated scores are at least 2 points, which corresponds to a sufficiently high activity.

Results of testing new compounds on isolated tissue model
The results of the study of A 1 -antagonistic activity of 6 nitro derivatives of tetrazolo[1,5-a]pyrimidine and 1 derivative of 1,2,4-triazolo[1,5-a]pyrimidine are shown in Table 2.
Only one of the seven tested compounds was characterized by an A 1 -blocking effect in vitro at a concentration of 10 μmol/L: VIII. It is confirmed by more than 45% suppression relative to the control measurements, and, compared to the reference drug caffeine, A 1 -blocking effect of compounds was inferior by 1.3 times. Compounds XII and IX did not show any A 1 -antagonistic activity. Derivative of 1,2,4-triazolo[1,5-a]pyrimidine III was inferior to caffeine in terms of its antagonistic activity against A 1 -adenosine receptors by 1.5 times.

Conclusion
In the course of the study, in silico prediction of adenosine A 1 receptor affinity of 25 new compounds was made using the method of maximum structural similarity to the reference standards. As a result of the prediction, 7 promising compounds with the expected moderate or high A 1 -antagonistic activity were identified, which were recommended for experimental studies. Of these, three   compounds were recommended to be tested for their A 1 -antagonistic activity as a priority.
Seven new nitro derivatives of azolo[1,5-a]pyrimidines were tested for A 1 -antagonistic activity in vitro. The results of in silico prediction of activity of the compounds under study were confirmed for compound VIII. In terms of isolated atria technique, the inhibition of adenosine negative chronotropic effect was about 45%, under the influence of this substance. So, compound VIII is of interest for us for further in-depth study of its pharmacological properties.

Conflict of interests
The authors declare no conflict of interests.  [PubMed]