The Specificity and Broad Multitarget Properties of Ligands for the Free Fatty Acid Receptors FFA3/GPR41 and FFA2/GPR43 and the Related Hydroxycarboxylic Acid Receptor HCA2/GPR109A

The paradigm of ligand-receptor interactions postulated as “one compound—one target” has been evolving; a multi-target, pleiotropic approach is now considered to be realistic. Novel series of 1,4,5,6,7,8-hexahydro-5-oxoquinolines, pyranopyrimidines and S-alkyl derivatives of pyranopyrimidines have been synthesized in order to characterise their pleiotropic, multitarget activity on the FFA3/GPR41, FFA2/GPR43, and HCA2/GPR109A receptors. Hexahydroquinoline derivatives have been known to exhibit characteristic activity as FFA3/GPR41 ligands, but during this study we observed their impact on FFA2/GPR43 and HCA2/GPR109A receptors as well as their electron-donating activity. Oxopyranopyrimidine and thioxopyranopyrimidine type compounds have been studied as ligands of the HCA2/GPR109A receptor; nevertheless, they exhibited equal or higher activity towards FFA3/GPR41 and FFA2/GPR43 receptors. S-Alkyl derivatives of pyranopyrimidines that have not yet been studied as ligands of GPCRs were more active towards HCA2/GPR109A and FFA3/GPR41 receptors than towards FFA2/GPR43. Representative compounds from each synthesized series were able to decrease the lipopolysaccharide-induced gene expression and secretion of proinflammatory cytokines (IL-6, TNF-α) and of a chemokine (MCP-1) in THP-1 macrophages, resembling the effect of HCA2/GPR109A ligand niacin and the endogenous ligand propionate. This study revealed groups of compounds possessing multitarget activity towards several receptors. The obtained data could be useful for further development of multitarget ligands.


Introduction
The paradigm of ligand-receptor interactions postulated as "one compound-one target" has been transformed over the years by the recognition that a multi-target, pleiotropic approach can also be remarkably productive or even preferable [1].
G-protein-coupled receptors (GPCRs) provide attractive targets for investigating the putative pleiotropy of ligands. GPCRs, with more than 800 members, constitute the largest family of membrane proteins that are widely expressed in the human body and are involved in the regulation of most physiological functions through the modulation of cell signaling pathways via G-proteins or β-arrestin. Consequently, GPCRs are important targets of many synthetic ligands. At least 108 GPCRs serve as targets for 475 (34%) of the drugs approved by United States Food and Drug Administration, and 68 potentially novel receptors are targeted by 320 (20%) of the drug candidates that are in clinical trials [2]. At the same time, the overwhelming majority of GPCRs (~220) still have not been successfully targeted by such small molecules as ligands.
The free fatty acid receptors 2 and 3 (FFA2/GPR43 and FFA3/GPR41) and the hydroxycarboxylic acid receptor 2 (HCA2/GPR109A) belong to the class of GPCRs that are activated by the products of dietary fiber digestion: metabolic intermediates such as short-chain fatty acids (acetate, propionate, butyrate) and hydroxylated carboxylic acids [3,4].
The FFA and HCA receptors are coupled to the pertussis toxin-sensitive, G i/o -alpha family of G-proteins, and FFA2 is also coupled to the G q family. Their activation results in the inhibition of intracellular cAMP accumulation and mobilisation of intracellular calcium, respectively. These receptors are coexpressed in intestinal epithelial cells, pancreatic βcells, adipocytes, and immune cells such as neutrophils, monocytes, and intestinal Treg cells, and exhibit some overlap in agonist (butyrate) specificity [5,6]. They are involved in multiple regulatory pathways affecting several important biological processes, such as gastrointestinal and immune functions and the regulation of energy homeostasis.
The FFA2/GPR43 and FFA3/GPR41 receptors are involved in energy homeostasis via the modulation of hormones, such as glucagon-like peptide 1, gastric inhibitory peptide, and peptide YY, affecting secretion from intestinal endocrine cells, suppression of fat accumulation in adipose tissue, and leading to increased insulin sensitivity [5,7]. The FFA3/GPR41 receptor also enhances insulin sensitivity through the gut-brain neural circuit, involving the activation of peripheral nerves [8]. The protective effects of propionate on the blood-brain barrier have been shown to occur by binding to FFA3/GPR41 on the surface of human brain endothelial cells, leading to a reduction in inflammatory and oxidative processes [9]. The FFA2/GPR43, FFA3/GPR41, and HCA2/GPR109A ligands reduce colonic inflammation by regulating the production of cytokines and chemokines by immune system cells. This mechanism has been shown to affect inflammatory responses in models of asthma, colitis, and arthritis [10][11][12].
Therefore, FFARs are regarded as novel drug targets for the treatment of metabolic disorders such as obesity, type 2 diabetes, gastrointestinal disorders, colorectal cancer, and diseases of the immune system [13][14][15].
There have been numerous attempts to approach the problem of obtaining selective and effective ligands for modulating biochemical pathways that involve metabolite GPCRs [2,[16][17][18][19][20][21][22]. According to our knowledge, there has been no comparative study of the activities of putative ligands on these three GPCRs, namely, the free fatty acid receptors FFA2/GPR43 and FFA3/GPR41 and the hydroxycarboxylic acid receptor HCA2/GPR109A. A multitarget, pleiotropic approach can in principle enable the discovery of pharmaceutical agents combining the therapeutic effects of two or three separate drugs. Such an achievement could reduce the range of side effects, lower the costs of drug development and manufacturing, and possibly shorten drug development times. Already, 33% (156) of approved drugs targeting GPCRs have more than one therapeutic indication [23,24].
Rationally designed polypharmacological studies using a set of ligands or multi-target directed ligands have been proposed as the only approach that can confront the complexity of multifactorial diseases, e.g., Alzheimer's disease [25][26][27][28][29].
The aim of this study was to determine the selectivity and multitarget characteristics of synthetic ligands for free fatty acid receptors (such as FFA2/GPR43 and FFA3/GPR41) and hydroxycarboxylic acid receptors.
It would be useful to test the activity of hydroxycarboxylic acid receptor ligands on the free fatty acid receptors FFA3/GPR41 and FFA2/GPR43 and, vice versa, the activity of shortchain fatty acid receptor ligands on the hydroxycarboxylic acid receptor HCA2/GPR109A. As a result of this cross-examination, common structural patterns of hydroxycarboxylic acid receptor ligands and short-chain fatty acid receptor ligands may be revealed, leading to a broader scope of applications for such compounds.
The aim of this study was to determine the selectivity and multitarget characteristics of synthetic ligands for free fatty acid receptors (such as FFA2/GPR43 and FFA3/GPR41) and hydroxycarboxylic acid receptors.
It would be useful to test the activity of hydroxycarboxylic acid receptor ligands on the free fatty acid receptors FFA3/GPR41 and FFA2/GPR43 and, vice versa, the activity of short-chain fatty acid receptor ligands on the hydroxycarboxylic acid receptor HCA2/GPR109A. As a result of this cross-examination, common structural patterns of hydroxycarboxylic acid receptor ligands and short-chain fatty acid receptor ligands may be revealed, leading to a broader scope of applications for such compounds.
The in vivo biological activity of many compounds is associated with their participation in and the effect on free radical reactions. Therefore, we evaluated the electron-donating properties of our synthesized compounds by their electro-oxidation potentials.
The FFA receptor ligands mediate their protective effect on human brain endothelial cells through targeting the blood-brain barrier. For this reason, we calculated the topological polar surface area (TPSA) which helped us to predict the blood-brain barrier penetration ability for each compound [32]. We used calculated log p values to evaluate the lipophilicity of the studied compounds.
Our study has allowed us to characterise the similarity in the action by test compounds on the short-chain fatty acid receptors FFA2/GPR43, FFA3/GPR41, and to provide a comparison of their activity towards the hydroxycarboxylic acid receptor HCA2/GPR109A. Hexahydroquinoline derivatives were mainly selective and active on the FFA3/GPR41 receptor in the low micromolar range, but weaker activity towards the FFA2/GPR43 and/or HCA2/GPR109A receptors can also be manifested. The investigated oxo-and thioxopyranopyrimidines not only served as HCA2/GPR109A ligands but also The HCA2/GPR109A, FFAR3/GPR41, and FFA2/GPR43 receptors are co-expressed in different types of immune cells, such as monocytes, macrophages, and neutrophils, where they reduce inflammatory function by inhibiting proinflammatory cytokine secretion and chemotaxis [30,31]. Therefore, we studied putative effects of compounds belonging to the groups 1, 2 and 3 on the proinflammatory cytokine levels in macrophages.
The in vivo biological activity of many compounds is associated with their participation in and the effect on free radical reactions. Therefore, we evaluated the electrondonating properties of our synthesized compounds by their electro-oxidation potentials.
The FFA receptor ligands mediate their protective effect on human brain endothelial cells through targeting the blood-brain barrier. For this reason, we calculated the topological polar surface area (TPSA) which helped us to predict the blood-brain barrier penetration ability for each compound [32]. We used calculated log p values to evaluate the lipophilicity of the studied compounds.
Our study has allowed us to characterise the similarity in the action by test compounds on the short-chain fatty acid receptors FFA2/GPR43, FFA3/GPR41, and to provide a comparison of their activity towards the hydroxycarboxylic acid receptor HCA2/GPR109A. Hexahydroquinoline derivatives were mainly selective and active on the FFA3/GPR41 receptor in the low micromolar range, but weaker activity towards the FFA2/GPR43 and/or HCA2/GPR109A receptors can also be manifested. The investigated oxo-and thioxopyranopyrimidines not only served as HCA2/GPR109A ligands but also stimulated the FFA3/GPR41 receptor, as well as the FFA2/GPR43 receptor with lower potency.
The studied compounds showed anti-inflammatory properties in macrophages by reducing the proinflammatory cytokine and chemokine levels; thus, an additional argument can be made that these ligands could possess pharmacological multitarget properties to treat such complex medical conditions as cancer, Alzheimer's disease, Parkinson's disease, and diabetes. The HHQ derivatives were prepared by three-or four-component Hantzsch-type cyclocondensation reactions according to the paths A, B or C (Scheme 1), using cyclohexane-Pharmaceuticals 2021, 14, 987 4 of 28 1,3-dione or 3-aminocyclohex-2-en-1-one, aryl-or heterocyclic carbaldehydes, and the appropriate 3-oxo-alkanoates or 3-oxo-carboxamides, or anilides.
Pyranopyrimidinetrione 2-5 was prepared according to a published method [19] by the condensation of barbituric acid with 4-alkyl-3-oxo-carboxylic acid esters. Known thioxopyranopyrimidines 2-1 and 2-2 were synthesized according to another reported procedure [20] using thiobarbituric acid instead of barbituric acid, as in the aforementioned method. The synthetic routes for compounds 2 and 3 are presented in Scheme 2.
Further alkylation of 2-thioxopyranopyrimidines 2 (X = S) was achieved through the piperidinium salts of compound 2 as non-isolated intermediates, using alkyl bromides as alkylating agents. Alkylthiopyranopyrimidinediones) 3-1-3-7 Pyranopyrimidinetrione 2-5 was prepared according to a published method [19] by the condensation of barbituric acid with 4-alkyl-3-oxo-carboxylic acid esters. Known thioxopyranopyrimidines 2-1 and 2-2 were synthesized according to another reported procedure [20] using thiobarbituric acid instead of barbituric acid, as in the aforementioned method. The synthetic routes for compounds 2 and 3 are presented in Scheme 2. Further alkylation of 2-thioxopyranopyrimidines 2 (X = S) was achieved through the piperidinium salts of compound 2 as non-isolated intermediates, using alkyl bromides as alkylating agents.

The Impact of Oxo-Hexahydroquinolines on HCA2/GPR109A, FFA3/GPR41, and FFA2/GPR43 Receptors
The first group of compounds studied as potential ligands of the HCA2/GPR109A, FFA3/GPR41, and FFA2/GPR43 receptors were the hexahydroquinoline (HHQ) derivatives 1. For a long time, the data regarding HHQ derivatives as FFA3/GPR41 agonists and antagonists has been very limited [20,21]. Recently, a new paper has been published about structure-activity relationship studies of tetrahydroquinolones as FFA3 modulators, mainly focused on 4-alkyl-and 4-heteryl-HHQ 3-carboxamides [22]. Compound 1-3 (Table 1) has been synthesized by Arena Pharmaceuticals, evaluated as a selective agonist of the FFA3/GPR41 receptor [21,22], and proposed as a potential therapeutic agent for hepatocellular carcinoma. In order to compare our compounds to literature data, we re-synthesized three previously reported compounds (1-1, 1-2, 1-3) according to published procedures [21,22], motivated by the reported experimental data regarding their activity against one receptor (FFA3/GPR41). There are no data on these  The first group of compounds studied as potential ligands of the HCA2/GPR109A, FFA3/GPR41, and FFA2/GPR43 receptors were the hexahydroquinoline (HHQ) derivatives 1. For a long time, the data regarding HHQ derivatives as FFA3/GPR41 agonists and antagonists has been very limited [20,21]. Recently, a new paper has been published about structure-activity relationship studies of tetrahydroquinolones as FFA3 modulators, mainly focused on 4-alkyl-and 4-heteryl-HHQ 3-carboxamides [22]. Compound 1-3 (Table 1) has been synthesized by Arena Pharmaceuticals, evaluated as a selective agonist of the FFA3/GPR41 receptor [21,22], and proposed as a potential therapeutic agent for hepatocellular carcinoma. In order to compare our compounds to literature data, we re-synthesized three previously reported compounds (1-1, 1-2, 1-3) according to published procedures [21,22], motivated by the reported experimental data regarding their activity against one receptor (FFA3/GPR41). There are no data on these compounds regarding the FFA2/GPR43 and HCA2/GPR109A receptors. The HHQ derivatives 1-11, 1-12, and 1-14 are known compounds that have been synthesized as putative tools for inducing the degradation of transforming growth factor-beta receptor type II [33] (thus, a quite different type of biological activity) and were obtained according to the described procedures.
Compound 1-13 was synthesized as one of several early examples of HHQ derivatives [34], without any data on biological activity. Thus, practically no comparative data are available about the biological activity of HHQ derivatives as potential ligands of the free fatty acid receptors FFA3/GPR41 and FFA2/GPR43 and the hydroxycarboxylic acid receptor HCA2/GPR109A.
All of our synthesized compounds were evaluated in forskolin-stimulated cAMP accumulation assay using Flp-In-293 cells stably expressing human FFA2/GPR43, FFA3/GPR41 or HCA2/GPR109A receptors (Tables 1-3; Figure 2). The activity was described as EC 50 values for high potency compounds and as inhibition (%) of forskolin-stimulated intracellular accumulation of cAMP by 50 µM of tested compounds in cases of lower potency. compounds regarding the FFA2/GPR43 and HCA2/GPR109A receptors. The HHQ derivatives 1-11, 1-12, and 1-14 are known compounds that have been synthesized as putative tools for inducing the degradation of transforming growth factor-beta receptor type II [33] (thus, a quite different type of biological activity) and were obtained according to the described procedures. compounds regarding the FFA2/GPR43 and HCA2/GPR109A receptors. The HHQ derivatives 1-11, 1-12, and 1-14 are known compounds that have been synthesized as putative tools for inducing the degradation of transforming growth factor-beta receptor type II [33] (thus, a quite different type of biological activity) and were obtained according to the described procedures. compounds regarding the FFA2/GPR43 and HCA2/GPR109A receptors. The HHQ derivatives 1-11, 1-12, and 1-14 are known compounds that have been synthesized as putative tools for inducing the degradation of transforming growth factor-beta receptor type II [33] (thus, a quite different type of biological activity) and were obtained according to the described procedures. compounds regarding the FFA2/GPR43 and HCA2/GPR109A receptors. The HHQ derivatives 1-11, 1-12, and 1-14 are known compounds that have been synthesized as putative tools for inducing the degradation of transforming growth factor-beta receptor type II [33] (thus, a quite different type of biological activity) and were obtained according to the described procedures.  The calculated logP values showed moderate lipophilicity. Almost all compound had logP values below five (similarly to [22]) and TPSA less than 90: see comp. 1-1 and 4 [22]; comp. 1-2 and 1 [22]; comp. 1-3 and 16 [22]. Some tetrahydroquinolone compounds have been described as allosteric modulator of the FFA3/GPR41 receptor that can increase or decrease the potency of endogenou ligand propionate [22]. Therefore, the ability of compounds 1-1, 1-2, 1-3, and 1-4 to decrease forskolin-stimulated cAMP levels was tested in the presence or absence of 5 µM propionate. The potency of compound 1-2 in the absence or presence of propionate wa the same (EC50 = 0.61 ± 0.09 µM and EC50 = 0.58 ± 0.03 µM, respectively), which was in a good agreement with the potency reported by Ulven et al. [22] for compound 1 (EC50 = 0.52 µM and EC50 = 0.54 µM, respectively). The potency of compound 1-1 (EC50 = 1.79 ± 0.3 µM) was slightly increased in the presence of propionate (EC50 = 0.64 ± 0.2 µM), with values close to those reported for compound 4 [23]. The potencies of compounds 1-3 (EC5 = 0.32 ± 0.05 µM) and 1-4 (EC50 = 0.23 ± 0.07 µM) were increased in the presence o propionate (EC50 = 0.10 ± 0.02 µM and EC50 = 0.081± 0.04 µM, respectively). These result suggest that compounds 1-1, 1-3, and 1-4 could stimulate the FFA3/GPR41 receptor a agonists and potentiate the activity of propionate.

The Effects of Pyranopyrimidine Derivatives on HCA2/GPR109A, FFA2/GPR43, and FFA3/GPR41 Receptors
Pyranopyrimidines have been tested so far only as HCA2/GPR109A receptor ligand [19,20]. We studied the possible activities of pyranopyrimidines towards three GPCRs HCA2/GPR109A, FFA2/GPR43, and FFA3/GPR41. Besides the newly synthesized compounds, we studied two known compounds (2-2 and 2-5) for which literature data were available concerning their activity towards one receptor (HCA2/GPR109A). Some structure-activity relationships of the studied compounds (Table 2) have been identified We observed two main types of ligand-receptor potency patterns. This activity pattern can be divided into subtypes: in four cases, the compounds were Compounds were added at the indicated concentrations to measure their inhibition of forskolin-stimulated cAMP production. The curve represents the mean ± SD from two independent experiments performed in triplicate.
Modified compounds 1-11 and 1-14, featuring carboxylate ester groups at position 3, had low to no activity towards the FFA3/GPR41 receptor. Compound 1-8, possessing a morpholide group, showed a low impact on FFA3/GPR41 and lacked activity towards the FFA2/GPR43 and HCA2/GPR109A receptors. Compound 1-12, with an N-alkyl lactam ring instead of an alkylamide group, had low activity towards all three studied receptors. Thus, several compounds exhibited comparable activities towards all three receptors, such as compound 1-11 (having a propylaminomethyl group at position 2 of the hexahydroquinoline system), as well as compounds 1-12 and 1-13. Compound 1-14 (featuring a carboxyl group at position 2 of the HHQ system) completely lacked activity towards all three receptors. All three compounds (1-11, 1-12, 1-14) had a biphenyl system at position 4; the same system is also part of compounds 5 and 7 [22], and the combination of a biphenyl substituent with o-toluidide functionality led to the highest activity for the FFA3/GPR41 receptor in the studied set of compounds. At the same time, hexahydroquinoline-3-carboxylic acid anilide (compound 1-1) lacked activity towards FFA2/GPR43. On the other hand, hexahydroquinoline-3-carboxylic acid esters (compounds 1-9 and 1-11) showed activity towards FFA2/GPR43 and HCA2/GPR109A. Compound 1-10 had reduced activity towards FFA3/GPR41 compared to compound 1-9 due to the additional phenyl group, but it had additional low activity towards the HCA2/GPR109A receptor.

The Effects of Pyranopyrimidine Derivatives on HCA2/GPR109A, FFA2/GPR43, and FFA3/GPR41 Receptors
Pyranopyrimidines have been tested so far only as HCA2/GPR109A receptor ligands [19,20]. We studied the possible activities of pyranopyrimidines towards three GPCRs: HCA2/GPR109A, FFA2/GPR43, and FFA3/GPR41. Besides the newly synthesized compounds, we studied two known compounds (2-2 and 2-5) for which literature data were available concerning their activity towards one receptor (HCA2/GPR109A). Some structure-activity relationships of the studied compounds (Table 2) have been identified. We observed two main types of ligand-receptor potency patterns.
This activity pattern can be divided into subtypes: in four cases, the compounds were more active towards HCA2/GPR109A, but in one case compound 2-4 (a derivative of 2-thioxo-5-butylpyranopyrimidine) was more active towards FFA3/GPR41 than HCA2/GPR109A. Thus, in principle, either multitarget or selectively active compounds possessing different selectivity could be designed. The prolongation of the alkyl chain (propyl → butyl) enhanced activity towards all studied receptors (according to the comparison of compounds 2-3 and 2-4). We added a novel 2-oxopyranopyrimidine derivative to the group of compounds described by Palani et al. [18,19,35]: 5-chlorobutyl-2-oxopyranopyrimidine (compound 2-7). It had the same activity on the HCA2/GPR109A receptor as its butyl analogue, but the impact on FFA3/GPR41 and FFA2/GPR43 was significantly increased compared to the 5-butyl compound 2-5.
Evidently, compounds of this type provide an appropriate platform for further optimization in the most preferable direction.
The observed potency of compound 2-3 matched literature data [19], but there was a difference in the case of compound 2-5 [18].
2.4. The Impact of S-Alkylpyranopyrimidine Derivatives on HCA2/GPR109A, FFA3/GPR41, and FFA2/GPR43 Receptors S-Alkylpyranopyrimidine derivatives had not yet been studied as ligands of shortchain fatty acid or hydroxycarboxylic acid receptors. Remarkably, the aforementioned compounds were shown to have activity towards all three receptors, especially the FFA3/GPR41 and HCA2/GPR109A receptors (Table 3).
2-Benzyloxybenzoylethyl derivative 3-4: a compound possessing two phenyl rings and a ketone carbonyl group instead of an ester group had the highest potency towards FFA3/GPR41, but lower activities towards FFA2/GPR43 and HCA2/GPR109A.
The studied compounds had a remarkable potency (EC 50 from 0.38 to 8.2 µM for selected compounds) towards FFA3/GPR41 albeit not reaching the high potencies of the aforementioned oxo-or thioxopyranopyrimidines ( Table 2), but exceeding their potency towards HCA2/GPR109A (EC 50 from 17 to 75 µM or even lower concentration). In the case of activity towards the FFA2/GPR43 r, it was not possible to determine EC 50 , which was in some cases also true for FFA3/GPR41.
It can be concluded that S-alkylpyranopyrimidines are more selective for the FFA3/ GPR41 and HCA2/GPR109A receptors than for the FFA2/GPR43 receptor.
The activity of oxo-and thioxopyranopyrimidines towards the studied receptors was different (Table 2), as the mentioned ligands preferentially had more impact on HCA2/GPR109A receptor, but S-alkylpyranopyrimidines preferentially had more impact on FFA3/GPR41 receptor (Table 3). Such knowledge could be useful for further development of specific and multitarget ligands.
Compounds that were not able to change the forskolin-stimulated cAMP level were tested for their antagonist activity in the presence of the respective receptor agonist. None of these compounds were able to change the agonist activity (data not shown).
In general, it is possible to obtain compounds possessing high potencies on one, two or three receptors, or high selective activity on one out of three receptors. with the hexahydroquinoline type compounds 1-1, 1-2, 1-4, and 1-9.
Evidently, compounds of this type provide an appropriate platform for further optimization in the most preferable direction.
The observed potency of compound 2-3 matched literature data [19], but there was a difference in the case of compound 2-5 [18].  (comp. 3-2) led to a decrease of activity towards FFA3/GPR41, no change in activity towards FFA2/GPR43, but an increase in potency towards HCA2/GPR109A. 2-Benzyloxybenzoylethyl derivative 3-4: a compound possessing two phenyl rings and a ketone carbonyl group instead of an ester group had the highest potency towards FFA3/GPR41, but lower activities towards FFA2/GPR43 and HCA2/GPR109A. a ketone carbonyl group instead of an ester group had the highest potency towards FFA3/GPR41, but lower activities towards FFA2/GPR43 and HCA2/GPR109A. a ketone carbonyl group instead of an ester group had the highest potency towards FFA3/GPR41, but lower activities towards FFA2/GPR43 and HCA2/GPR109A. FFA3/GPR41, but lower activities towards FFA2/GPR43 and HCA2/GPR109A.
The antioxidant activity of dihydropyridines was subsequently correlated with different biological and pharmacological properties of these compounds [43,44]. The antioxidant activity of DHP derivatives was characterised using electrochemical methods [45].
Compounds containing fatty chains as substituents at the positions 2 or 3 and an aryl moiety at position 4 have good antioxidant activity, approaching even that of the classic antioxidants BHT (butylated hydroxytoluene) and vitamin E.

Lipophilicity and Polar Surface Area of Studied Compounds
The lipophilicity and polar surface area of the three studied groups of compounds were calculated as basic molecular characteristics pertaining to the potential medicinal use of these compounds.
Almost all of the synthesized and studied compounds had acceptable lipophilicity for peroral bioavailability (logP < 5) [49]. According to the calculated values, even lower lipophilicity was observed, thus a broader range of functionalities could be included in similar compounds during further structural optimization in order to design potential drug molecules for injectable final dosage forms [50].
Short-chain fatty acids have strong effects on neuronal physiology. The free fatty acid receptors FFA2/GPR43 and FFA/GPR41 are expressed in the brain [51,52] and play a significant role in human health conditions. The TPSA (topological polar surface area, Å 2 ) values characterise the ability of molecules to cross the blood-brain barrier. The calculated TPSA values for derivatives of hexahydroquinoline, oxopyranopyrimidines, and thioxopyranopyrimidines were estimated to mostly be lower than 90 Å 2 (Tables 1 and 2), while the calculated TPSA values for alkylmercaptopyranopyrimidines ranged from 97 to 127 Å 2 or even 135 Å 2 (Table 3), making these compounds in general appropriate for transport through cell membranes, and in most cases also through the blood-brain barrier [53].

Ligands of HCA2/GPR109A, FFA3/GPR41, and FFA2/GPR43 Receptors Suppress Gene Expression and Secretion of Inflammatory Cytokines in THP-1 Macrophages
Besides the studies focused on the activity of compounds on the free fatty acid receptors FFA3/GPR41 and FFA2/GPR43 and the hydroxycarboxylic acid receptor HCA2/GPR109A we took into consideration the co-expression of HCA2/GPR109A, FFA3/GPR41, and FFA2/GPR43 in immune cells (monocytes, macrophages, neutrophils), where they reduce proinflammatory function [54]. In order to investigate the biological activities of com-pounds 1, 2, 3, we stimulated human THP-1 macrophages with the test compounds at 10 µM and 50 µM concentrations, 50 µM of the HCA2/GPR109A selective ligand niacin, as well as 0.1 mM and 1 mM of the FFA2/GPR43 and FFAR3/GPR41 ligand propionate .  Compounds 1-1, 1-2 and 1-4 were tested as FFA3/GPR41 selective compounds, while  compounds 2-7, 3-3, and 3-4 were tested as compounds lacking selectivity for that receptor .  Compounds 1-1 and 1-2 are well-known representatives of the hexahydroquinoline group of FFAR3/GPR41 ligands, having an o-toluidide side chain at position 3 and 4-thienyl or 4-furyl group at position 4, and compound 1-4 is a novel active ligand of FFAR3/GPR41, possessing a 2,4-dichloroanilide substituent at position 3, representing an isomer of the known 2,5-dichloroanilide compound 1-3. Concerning the derivatives of pyranopyrimidines, we studied interesting 5-chlorobutyl derivatives: compound 2-7 as a representative of 2-oxopyranopyrimidines and two S-alkylthiopyranopyrimidines, 3-3 and 3-4, bearing ester and benzoylethyl moieties. First, we examined the gene expression of all three receptors in unstimulated and 4 h LPS-stimulated human THP-1 macrophages. Figure 3 shows that all three receptors were expressed in THP-1 cells, with higher HCA2/GPR109A levels than FFA2/GPR43 and FFA3/GPR41. LPS stimulation of cells markedly increased the gene expression for all three receptors compared to unstimulated cells. Niacin significantly reduced LPS upregulation of the HCA2/GPR109A (p < 0.01), but did not affect the expression of FFA2/GPR43 and FFA3/GPR41 genes. Compounds 2-7, 3-3 and 3-4 at 50 µM concentration also significantly reduced the LPS-stimulated HCA2/GPR109A mRNA levels (p < 0.05), resembling the effect of niacin. In contrast, the LPS-upregulated FFA2/GPR43 and FFA3/GPR41 gene expression was slightly increased by the addition of 1mM propionate and some of the compounds; however, this effect was not statistically significant compared to the LPS-stimulated cells (p > 0.05) (Figure 3). Propionate at 1 mM and compounds 1-1, 1-2, 2-7 and 3-3 at 50 µM concentrations significantly increased the LPS-stimulated FFA3/GPR41 mRNA levels (p < 0.05).  The gene expression for the receptors was analysed by real-time PCR and is reported relative to the value from untreated control cells, which was defined as 1. The data were normalized to RPS29 and the values were calculated using the comparative (2 −∆Ct ) method. The results are given as the mean ± SD from three independent experiments. * p < 0.05, ** p < 0.01 are significantly different from LPS-stimulated cells.
Our findings generally agree with the recent publications about multitarget pleiotropic drugs including 1,4-DHPs [55] and their usefulness as modulators for GPCR (especially anti-inflammatory SCFA receptors) for the treatment of Diabetes mellitus (T1D) [56]. receptors and the nuclear factor κB (NF-κB) pathway, with subsequent regulation of inflammatory genes. It has been shown that the FFA and HCA2 receptors mediated signaling pathways suppress NF-κB signalling in lung tissues and macrophages [6,12]. The observed inhibition of proinflammatory cytokines in macrophages by the studied compounds can be mediated via inactivation of NF-κB; however, the precise mechanism should be further elucidated. Our findings generally agree with the recent publications about multitarget pleiotropic drugs including 1,4-DHPs [55] and their usefulness as modulators for GPCR (especially anti-inflammatory SCFA receptors) for the treatment of Diabetes mellitus (T1D) [56].

Materials
All materials for cell cultures were supplied by Thermo Fisher Scientific. A Lance cAMP kit was purchased from PerkinElmer. The pcDNA3.1-FFA2R and pcDNA3.1-FFA3R plasmids were obtained from cDNA Resource Center (www.cdna.org; accessed on 20 June 2020). All other compounds and reagents were purchased from Acros Organics, Sigma-Aldrich or Alfa Aesar, unless stated otherwise, and were used without further purification.

Materials
All materials for cell cultures were supplied by Thermo Fisher Scientific. A Lance cAMP kit was purchased from PerkinElmer. The pcDNA3.1-FFA2R and pcDNA3.1-FFA3R plasmids were obtained from cDNA Resource Center (www.cdna.org; accessed on 20 June 2020). All other compounds and reagents were purchased from Acros Organics, Sigma-Aldrich or Alfa Aesar, unless stated otherwise, and were used without further purification.
(b) A mixture of cyclohexane-1,3-dione (0.56 g, 5 mol), the appropriate aldehyde (5 mmol), the appropriate alkyl 3-oxobutanoate or anilide (5 mmol), and ammonium acetate (1.2 g, 15 mmol) in ethanol (15 mL) was treated by adding five drops of acetic acid, followed by stirring at rt for 24 h. The precipitate that formed upon cooling in a refrigerator was collected by filtration, then washed with water and cold ethanol to afford compounds 1-8, 1-9.
The precipitate was filtered and washed with hot ethyl acetate (10 mL). The solution was concentrated in vacuo. The residue was dissolved in hot DMF (3 mL); water (~4 mL) was then added dropwise until the precipitation was complete. The light grey powder was collected by filtration, recrystallized from ethanol, and dried at reduced pressure, yielding compound 2-7 (210 mg, 11%), Mp > 200 • C. 1  A mixture of thiobarbituric acid (1.44 g, 10 mmol, 1 eq) and the appropriate ethyl (methyl) 4-alkyl-3-oxo-alkanoate (12.6 mmol, 1.26 eq) was added to a vial with glacial acetic acid (15 mL) under argon atmosphere and the mixture was heated at 120 • C for 30 h. The hot mixture was filtered and the solids were washed with hot acetic acid. The acetic acid was removed from the filtrate under reduced pressure and the residue was triturated with hot ethyl acetate (70 mL). The precipitate was collected by filtration and washed with hot ethyl acetate (10 mL). The solution was concentrated under reduced pressure. The residue was dissolved in hot DMF (4 mL) and then water (~5 mL) was added dropwise until the precipitation was complete. The obtained light grey powder was collected by filtration, recrystallized from ethanol, and dried under reduced pressure. The obtained 2-thioxopyranopyrimidines were recrystallized from ethanol. A mixture of the appropriate 2-thioxo-pyranopyrimidine 2 (1 mmol, 1 eq) and piperidine (0.104 mL, 1.01 mmol) in ethanol (5 mL) was heated to reflux over~10 min until the colour of the solution changed. The appropriate alkyl(aralkyl)bromide (1 mmol, 1 eq) was then added and the solution was refluxed for 10 min and stirred at rt for 24 h. After cooling, the precipitate was collected by filtration and washed with cold ethanol and water. The crude product was recrystallized from ethanol and dried under reduced pressure. medium was exchanged for PMA-free medium for 48 h before the cells were used for experiments. For studying the effects of ligands on cytokine and receptor gene expression and cytokine secretion, the differentiated cells were stimulated for 4 h with 1 µg/mL LPS in the presence or absence of propionate (0.1 mM and 1 mM), niacin (50 µM), and the studied compounds (10 µM and 50 µM).

Intracellular cAMP Inhibition Assay
For cAMP assay, the cells were distributed into a 384-well white microplate at a density of 12,000 cells per well and stimulated with the indicated compounds in the presence of 3 µM forskolin for 30 min. Compounds (50 µM) that were not active in the cAMP assay were tested for their antagonist activity in the presence of the respective receptor agonist: 5 µM niacin (HCA2/GPR109A), 1 µM 1-3 (FFA3/GPR41) or 1 µM 2-7 (FFA2/GRP43). The cells and the compounds were diluted in 1X phosphate-buffered saline (PBS) supplemented with 0.5 mM of 3-isobutyl-1-methylxanthine. The intracellular cAMP levels were measured using a Lance cAMP kit according to the manufacturer's instructions. The plates were read on a PerkinElmer Victor 3V multilabel plate reader. The cAMP was quantitated in each sample by comparison to a calibration curve of known cAMP quantities provided in the kit.

Quantification of Cytokines and Chemokines
Conditioned media were obtained from untreated and compound-treated THP-1 macrophages. The concentrations of the IL-6, TNF-α and MCP-1 were measured by Luminex Multiplex immunoassay (R&D Systems, USA) according to the manufacturer's instructions, and analyzed on a Luminex 2000 (Luminex Corporation, Austin, TX, USA). Three independent repetitions in duplicate were made per sample. Concentrations were quantified with five parameters logistic (5-PL) curve fit and expressed in pg/mL.

Statistical Analysis
The gene expression results are presented as the mean values ± SD for the indicated number of experiments. Student's t-test was used to assess the statistical significance of differences. Significant differences were assumed for p < 0.05. The statistical and cAMP assay data analyses were performed using GraphPad Prism 5.0 software (GraphPad Software, San Diego, CA, USA).

Characterisation of the Synthesized Compounds
One-dimensional 1 H and 13 C NMR spectra were recorded on a Varian 400 Mercury spectrometer at 400 MHz (for 1 H nuclei) and 100 MHz (for 13 C nuclei). The chemical shifts are presented in parts per million (ppm). Residual protons of deuterated solvents were used as internal standard for 1 H NMR spectra (CDCl 3 : δ 7.26 ppm, DMSO-d 6 : δ 2.50). The deuterated solvent signals were used as internal standards for 13 C NMR spectra.
Low-resolution mass spectra (MS) were recorded on a Waters ACQUITY UPLC system equipped with a BEH C 18 column connected to a Waters SQ Detector 2 operating in the ESI positive ion mode. Elemental analyses were performed on an Elemental Combustion System ECS 4010 (Costech Instruments) at the Laboratory of Chromatography, Latvian Institute of Organic Synthesis. All target compounds had >95% purity. The purity of each compound was determined by HPLC on a Waters Alliance 2695 system equipped with an Altima C 18 column, 5 µM, 4.6 × 150 mm and a Waters 2489 UV-VIS detector, using a gradient elution with acetonitrile/H 3 PO 4 (0.1%) in water at a flow rate of 1 mL/min.

The Evaluation of Redox Potentials
Cyclic voltammetry experiments were carried out using a PARSTAT 2273 electrochemical system. A stationary glassy carbon disk electrode (d = 0.8 mm) served as the working electrode, while the counter electrode was a Pt wire. The oxidation potentials were measured relative to an Ag/Ag + reference electrode. Acetonitrile was dried over P 2 O 5 and distilled; the distillate was stored over CaH 2 and redistilled immediately prior to use. Recrystallized tetrabutylammonium tetrafluoroborate (TBATFB) was used as a supporting electrolyte at 0.1 M concentration.

The Calculation of logP and TPSA
The logP (logP OW ) and TPSA values were calculated with Chem3D Ultra 19 software (Perkin Elmer Informatics, Waltham, MA, USA). For the logP OW calculations, a Molecular Networks module was used on molecular species, while salts were treated as cations.