Triterpenic Acid Amides as a Promising Agent for Treatment of Metabolic Syndrome

: A series of triterpenic acid amides were synthesized incorporating a 2-ethoxy-3-phenylpropanoic acid pharmacophore fragment. The synthesized compounds were tested for their ability to improve glycemic control and to counter lipid abnormalities in C57BL/6 mice placed on a high-fat/high-cholesterol diet. Of all tested compounds, the dihydrobetulonic derivative ( 16b ) had the most pronounced effect in decreasing blood glucose levels, total cholesterol (TC), and high-density lipoproteins (HDL). All the synthesized compounds displayed a relatively safe proﬁle in the animal studies carried out in this work.


Introduction
Type 2 diabetes or diabetes mellitus is a complex disease with pleiotropic clinical manifestations directly connected to a more general cluster of conditions referred to as metabolic syndrome. Diabetic dysfunctions, characterized by hyperglycemia, result from a combination of factors: insulin resistance, inadequate insulin secretion, and excessive or inappropriate glucagon secretion. In the last two decades, a new class of promising anti-diabetic pharmacological targets has emerged, comprised of a subfamily of nuclear receptors activated by peroxisome proliferators (PPARs). The activation of these receptors has been shown to normalize metabolic dysfunctions and reduce cardiovascular risk factors associated with type 2 diabetes [1]. Among the different activators, dual PPAR-α,γ agonists (glitazars) are of particular interest for the treatment of metabolic syndrome since they combine the hypolipidemic and hypoglycemic properties of α and γ-agonists [2].
A number of dual PPAR agonists, demonstrating promising results in animal studies, have been tested in clinics. However, to date, only one, saroglitazar (Figure 1), has been approved in India [3]. All the others-ragaglitazar, muraglitazar, tesaglitazar, aleglitazar ( Figure 1)-failed to gain regulatory approval, primarily due to adverse side effects, that include hepatotoxicity, cardiotoxicity, and gastrointestinal toxicity [4,5]. Glitazars' adverse side effects may in principle be reduced by chemically modifying their pharmacophore groups. Among the possible modifications, the introduction of fragments derived from natural compounds is of particular interest.
Natural compounds are a well-known starting point for the synthesis of many drugs [6] and the introduction of natural pharmacophores can significantly improve their inherent properties. In this work, we focused on triterpenic acids and their derivatives, such as ursolic, corosolic, glycerrhetinic, as well as betulonic acids and their lipophilic derivatives. Lipophilic fragments are of particular interest in the case of PPARs because PPARs are actively expressed in the liver and in adipose tissues [7]. Triterpenic acids and their derivatives have both hepatoprotective and anti-diabetic properties [8][9][10][11], while being at the same time of little toxicity [12]. These favorable characteristics may be explained by the presence of a steroid nucleus, thought to be important for strong hypoglycemic effects [13] and their significant lipophilicity.
The triterpenic acids ( Figure 2) chosen in this work are known to have hypoglycemic properties. Thus, certain lupane-type triterpenoids, such as betulonic 6a and dihydrobetulonic 6b acids, have been shown to possess alpha-glucosidase inhibitory activities [14,15]. Glycyrrhetinic acid 7a is known to have several targets for its hypoglycemic actions: PPAR-γ, C/EBP-α, MAPK activation and pHSL [16]. In the case of ursolic acid 8a, the validated targets include PPAR-α, SREBP-1c, HSL translocation, perilipin A expression, ATGL and others. Glycyrrhetinic acid derivatives have been reported to possess therapeutic antidiabetic potential [17]. Of particular, recent interest is corosolic acid 8b that has a pronounced antidiabetic effect with a confirmed mechanism of AMPK activation and α-amylase inhibition [18]. Some derivatives of corosolic acid have also been shown to have pronounced hypoglycemic activity in vivo [19]. Glitazars' adverse side effects may in principle be reduced by chemically modifying their pharmacophore groups. Among the possible modifications, the introduction of fragments derived from natural compounds is of particular interest.
Natural compounds are a well-known starting point for the synthesis of many drugs [6] and the introduction of natural pharmacophores can significantly improve their inherent properties. In this work, we focused on triterpenic acids and their derivatives, such as ursolic, corosolic, glycerrhetinic, as well as betulonic acids and their lipophilic derivatives. Lipophilic fragments are of particular interest in the case of PPARs because PPARs are actively expressed in the liver and in adipose tissues [7]. Triterpenic acids and their derivatives have both hepatoprotective and anti-diabetic properties [8][9][10][11], while being at the same time of little toxicity [12]. These favorable characteristics may be explained by the presence of a steroid nucleus, thought to be important for strong hypoglycemic effects [13] and their significant lipophilicity.
The triterpenic acids ( Figure 2) chosen in this work are known to have hypoglycemic properties. Thus, certain lupane-type triterpenoids, such as betulonic 6a and dihydrobetulonic 6b acids, have been shown to possess alpha-glucosidase inhibitory activities [14,15]. Glycyrrhetinic acid 7a is known to have several targets for its hypoglycemic actions: PPARγ, C/EBP-α, MAPK activation and pHSL [16]. In the case of ursolic acid 8a, the validated targets include PPAR-α, SREBP-1c, HSL translocation, perilipin A expression, ATGL and others. Glycyrrhetinic acid derivatives have been reported to possess therapeutic antidiabetic potential [17]. Of particular, recent interest is corosolic acid 8b that has a pronounced antidiabetic effect with a confirmed mechanism of AMPK activation and α-amylase inhibition [18]. Some derivatives of corosolic acid have also been shown to have pronounced hypoglycemic activity in vivo [19]. Sci  As the basis of the synthesis of new compounds, we have chosen the 2-ethoxy-3-phenylpropanoic acid pharmacophoric fragment, a common structural fragment for most of the glitazars tested in clinics. The newly synthesized compounds were tested for their ability to improve glycemic control and to counter lipid abnormalities in vivo, in C57/BL/6 mice placed on a high fat diet with the addition of cholesterol and cholate.
Our approach to the fine-tuning of the pharmacological activity consisted in the modification of the glitazars pharmacophore fragment at the level of the phenolic hydroxyl group in the tyrosol fragment, as exemplified by tesaglitazar. We hypothesized that, just like with other existing glitazars, modifications in this portion of the molecule would not affect its binding to the receptor, but may lead to desirable pharmacokinetic and pharmacodynamic properties. The choice of amide bonding as a means to join the new pharmacophores is based on its known stability in biological environments, available protocols to create amide bonds and the possibility to incorporate numerous acids of natural origin. The presence of a spacer is important in two aspects. On the one hand, it provides some flexibility to the molecule, promoting enhanced binding to the receptor. On the other hand, the aminoethanol fragment constitutes a minimal requirement for the establishment of the amide bond linking the glitazars pharmacophore fragment at the level of the phenolic hydroxyl group. It should also be noted that corosolic acid on its own acts as a dual PPAR agonist. It is therefore of interest to study the influence of related terpene acids on the PPAR binding of their derivatives.

Chemistry
The analytical and spectral studies were conducted in the Chemical Service Center for the collective use of SB RAS.
The 1 H and 13 C-NMR spectra of the compounds in CDCl3 solutions were measured in a Bruker AV-400 spectrometer (400.13 and 100.61 MHz, respectively). The residual signals of the solvent were used as references (δH 2.48, δC 39.52 ppm for DMSO-d6 and δH 7.27, δC 77.1 ppm for CDCl3). Chemical shift measurements were calculated in ppm and the coupling constants (J) in hertz (Hz). The mass spectra (70 eV) were recorded on a DFS Thermo Scientific high-resolution mass spectrometer. Column chromatography employed Merck silica gel (63-200 μ). Thin-layer chromatography was performed on TLC Silica gel 60F254, Merck (Darmstadt, Germany). The melting temperatures of the compounds were determined using a Koffler table.
All chemicals were used as received unless otherwise noted. Reagent-grade solvents were redistilled prior to use. Synthetic starting materials, reagents and solvents were purchased from Sigma-Aldrich (St. Louis, USA), Acros Organics (Geel, Belgium) and AlfaAesar (Heysham, UK). As the basis of the synthesis of new compounds, we have chosen the 2-ethoxy-3phenylpropanoic acid pharmacophoric fragment, a common structural fragment for most of the glitazars tested in clinics. The newly synthesized compounds were tested for their ability to improve glycemic control and to counter lipid abnormalities in vivo, in C57/BL/6 mice placed on a high fat diet with the addition of cholesterol and cholate.
Our approach to the fine-tuning of the pharmacological activity consisted in the modification of the glitazars pharmacophore fragment at the level of the phenolic hydroxyl group in the tyrosol fragment, as exemplified by tesaglitazar. We hypothesized that, just like with other existing glitazars, modifications in this portion of the molecule would not affect its binding to the receptor, but may lead to desirable pharmacokinetic and pharmacodynamic properties. The choice of amide bonding as a means to join the new pharmacophores is based on its known stability in biological environments, available protocols to create amide bonds and the possibility to incorporate numerous acids of natural origin. The presence of a spacer is important in two aspects. On the one hand, it provides some flexibility to the molecule, promoting enhanced binding to the receptor. On the other hand, the aminoethanol fragment constitutes a minimal requirement for the establishment of the amide bond linking the glitazars pharmacophore fragment at the level of the phenolic hydroxyl group. It should also be noted that corosolic acid on its own acts as a dual PPAR agonist. It is therefore of interest to study the influence of related terpene acids on the PPAR binding of their derivatives.

Chemistry
The analytical and spectral studies were conducted in the Chemical Service Center for the collective use of SB RAS.
The 1 H and 13 C-NMR spectra of the compounds in CDCl 3 solutions were measured in a Bruker AV-400 spectrometer (400.13 and 100.61 MHz, respectively). The residual signals of the solvent were used as references (δH 2.48, δC 39.52 ppm for DMSO-d 6 and δH 7.27, δC 77.1 ppm for CDCl 3 ). Chemical shift measurements were calculated in ppm and the coupling constants (J) in hertz (Hz). The mass spectra (70 eV) were recorded on a DFS Thermo Scientific high-resolution mass spectrometer. Column chromatography employed Merck silica gel (63-200 µ). Thin-layer chromatography was performed on TLC Silica gel 60F254, Merck (Darmstadt, Germany). The melting temperatures of the compounds were determined using a Koffler table.
All chemicals were used as received unless otherwise noted. Reagent-grade solvents were redistilled prior to use. Synthetic starting materials, reagents and solvents were purchased from Sigma-Aldrich (St. Louis, MO, USA), Acros Organics (Geel, Belgium) and AlfaAesar (Heysham, UK).
Ursolic, betulonic and dihydrobetulonic acids were purchased from the Departmental Pilot of NIOCH SB RAS. Acetylglycyrrhetinic acid was donated by colleagues from the Department of Medicinal Chemistry at NIOCH SB RAS. Corosolic acid was synthesized from ursolic as described [20] and so was the reference compound tesaglitazar 3 [21]. The obtained spectral data coincide with those found in literature. (11) Tert-butyl 2-bromoethylcarbamate 10 3.42 g (15.3 mmol) and 4-(2-hydroxyethyl)phenol 9 1.51 g (11 mmol) were dissolved in 10 mL DMF in a 25-mL flask. K 2 CO 3 3.02 g (35 mmol) and catalytic amounts of TBAB were then added. The reaction was let to proceeded for 120 h with stirring in an inert atmosphere at 40 • C. The reaction was monitored by TLC in a CHCl 3

Oral Glucose Tolerance Test
The oral glucose tolerance test was carried out on the 32 day of the experiment, one day after the last administration of the compounds and after 12 h fasting. The same oral glucose load (2.5 g/kg) was used in all experimental groups. Blood glucose values were measured using a OneTouch Select blood glucose meter (LifeScan Inc., Milpitas, CA, USA) before dosing (0) and 30, 60, 90, 120 min after the glucose load. The area under the glycemic curve was calculated using Tai's model [22].

Biochemistry
After 5 weeks of treatment mice were decapitated and trunk blood was collected. The serum was isolated by centrifugation at 1640× g for 15 min. Serum total cholesterol (TC), total triglyceride (TG), high density and low-density lipoproteins (HDL and LDL), Glucose (Glu) and Alkaline phosphatase (ALP) levels were measured in all groups using standard diagnostic kits (Vector Best, Novosibirsk, Russia) and a Stat Fax 3300 spectrophotometer (Awareness Technology Inc., Palm City, FL, USA).
Statistical analysis was performed using Mann-Whitney U test. Results were calculated as averages ± SEM. p < 0.05 was considered to be statistically significant.

Chemistry
The synthetic route used for the preparation of the new triterpenic acid amides is presented in Scheme 1. Tert-butyl N-[2-[4-(2-hydroxyethyl)-phenoxy]-ethyl] carbamate 11 was obtained in 88% yield by the reaction of tyrosol 9 with an excess of bromide 10 in DMF in the presence of potassium carbonate and catalytic amounts of tetra-n-butylammonium bromide (TBAB). The conditions of the reaction are common with those of nucleophilic substitution. However, the elevated temperature at which the reaction took place led to a decrease in the yield of the target compound due to side reactions of bromide 10. At the same time, a decrease in the reaction temperature led to a significant increase in the reaction time. We found that 10 can be obtained in 88% yield when the reaction is carried out for seven days at 40 • C. reaction time. We found that 10 can be obtained in 88% yield when the reaction is carried out for seven days at 40 °C.
Free amine 14 was obtained in 92% yield using an adapted procedure [24] upon sequential treatment of trifluoroacetic acid on amine 13 with methylene chloride and an aqueous NaHCO3 solution. The condensation of this resulting amine 14 with acids 6-8 in the presence of (2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU) and N,N-diisopropylethylamine (DIPEA) in DMF using an adapted procedure [25] led to the production of novel amides 15a-f in yields of 65-89% under mild conditions. The resulting compounds 15a-f were hydrolyzed with lithium hydroxide using an adapted procedure [26] and the reaction mixture was subsequently acidified, leading to the formation of acids 16a-f in yields of 69-85%. Because the rate of hydrolysis of the acetyl group in the glycyrrhetinic moiety is much lower compared to the rate of hydrolysis of the ester group in the phenylpropanoic acid moiety, it is possible to isolate 16c in 85% yield as the main product when the reaction is carried out for approximately 2 h under cooling. Scheme 1. The synthetic route for synthesis of compounds 16a-f.
Free amine 14 was obtained in 92% yield using an adapted procedure [24] upon sequential treatment of trifluoroacetic acid on amine 13 with methylene chloride and an aqueous NaHCO 3 solution. The condensation of this resulting amine 14 with acids 6-8 in the presence of (2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU) and N,N-diisopropylethylamine (DIPEA) in DMF using an adapted procedure [25] led to the production of novel amides 15a-f in yields of 65-89% under mild conditions. The resulting compounds 15a-f were hydrolyzed with lithium hydroxide using an adapted procedure [26] and the reaction mixture was subsequently acidified, leading to the formation of acids 16a-f in yields of 69-85%. Because the rate of hydrolysis of the acetyl group in the glycyrrhetinic moiety is much lower compared to the rate of hydrolysis of the ester group in the phenylpropanoic acid moiety, it is possible to isolate 16c in 85% yield as the main product when the reaction is carried out for approximately 2 h under cooling.

Biology
To assess the pharmacological activities, the newly synthesized compounds were administered by oral gavage at a dose of 30 mg/kg for five weeks to mice placed on a high-fat diet with the addition of 2.5% cholesterol and 0.5% cholate (HF diet). This dose was determined according to preliminary experiments with similar molecules (data not published). Tesaglitazar 3 was used as a positive control at a dose of 15 mg/kg, to take into account its lower molecular weight. The changes in animal body weights were monitored throughout the experiment (Figure 3).

Biology
To assess the pharmacological activities, the newly synthesized compounds were administered by oral gavage at a dose of 30 mg/kg for five weeks to mice placed on a highfat diet with the addition of 2.5% cholesterol and 0.5% cholate (HF diet). This dose was determined according to preliminary experiments with similar molecules (data not published). Tesaglitazar 3 was used as a positive control at a dose of 15 mg/kg, to take into account its lower molecular weight. The changes in animal body weights were monitored throughout the experiment (Figure 3). week-period on HF-diet prior to introduction of testing compounds. Dosing starting point-first introduction of tesaglitazar, 16a, 16b, 16c, 16e, 16f. 16a-f were introduced at a dose of 30 mg/kg. An oral glucose tolerance test (OGTT) was carried out 32 days post administration of the tested compounds (Figures 4 and 5). The experiment was concluded after 36 days, and blood was drawn for biochemical analyses (Table 1).  tesaglitazar, 16a, 16b, 16c, 16e, 16f. 16a-f were introduced at a dose of 30 mg/kg. An oral glucose tolerance test (OGTT) was carried out 32 days post administration of the tested compounds (Figures 4 and 5). The experiment was concluded after 36 days, and blood was drawn for biochemical analyses (Table 1).   Based on the oral glucose tolerance test, the HF diet resulted in impaired glucose tolerance (HF diet control vs. SD after 60 min, Figure 4). A significant decrease in the concentration of blood glucose was observed only in the reference compound 3 and 16b groups. The lower values of their areas under the glycemic curve (AUC, Figure 5) do not differ from the values obtained in the control group (SD). Interestingly the greatest decrease in glucose was observed after 16b administration exceedingly early, only 60 min after the start of the experiment (Figure 4). Conversely, 16c administration caused an increase in blood glucose levels 60 and 90 min after glucose administration. This observation may be explained by a delay in insulin synthesis/release by pancreatic beta cells.
The mice maintained on the HF diet developed hyperlipidemia, as demonstrated by a significant increase in the blood levels of total cholesterol (TC) and carrier lipoproteins (HDL) ( Table 1). At the same time, the level of triglycerides (TG) was decreased, a finding that may be explained by a lower synthetic function of the liver and some hepatotoxic effects of the diet.
Administration of tesaglitazar 3 further reduced this parameter, while causing increased levels of TC and alkaline phosphatase (ALP). This finding can be considered a hepatotoxic effect since the introduction of other substances did not cause similar changes. Interestingly, the changes in biochemical parameters did not affect the body weight of the animals, which continued to gain weight throughout the experiment (Figure 3).
The desired hypolipidemic effect was found in mice treated with 16b. Their blood TC and HDL levels were significantly lower than those of the control group (HF diet). In addition, these animals also displayed the lowest glucose level, a finding that speaks in favor of a dual effect of 16b on lipid and carbohydrate metabolisms, absent in any of the other groups. ALP activities in these animals remained at the control level of mice maintained on both on HF and on SD diets. The absence of any ALP increase, which is associated with cholestasis, suggests that 16b is less toxic than reference compound 3. Administration of 16f and 16c led to a decrease in high density lipoproteins (HDL) cholesterol, without causing a significant decrease in TC, although the values are remarkably close to 16b. It is possible that these substances have a lower affinity for receptors than 16b and that a higher dose may be required to obtain a similar response.

Conclusions
A series of triterpenic acid amides were synthesized incorporating a 2-ethoxy-3phenylpropanoic acid pharmacophore fragment. The synthesized compounds were administered orally at a dose of 30 mg/kg for 5 weeks to mice placed on a high-fat/highcholesterol diet (HF diet). All the synthesized compounds, except for one (16c) appeared to have negligible toxic effects throughout the experiment. A significant decrease in blood glucose in the blood of animals, matching the levels in the control group (SD) was observed in animals receiving either the reference drug 3 or the dihydrobetulonic acid derivative 16b. The HF diet-associated increase in the blood of total cholesterol (TC) and carrier lipoproteins (HDL) and associated decrease in the level of triglycerides (TG) was effectively prevented by the administration of compound 16b, with full or partial normalization of these blood parameters in the relevant animals. Administration of tesaglitazar caused a further decrease in triglycerides (TG) levels and an increase in TC and alkaline phosphatase (ALP).
In conclusion, the newly synthesized dihydrobetulonic (16b) derivative is able to improve both lipid and carbohydrate metabolism and is thus a promising drug candidate, deserving further investigation.