The search for new drugs synthesis and antiviral activity of derivatives of meristotropic and macedonic acids

The antiviral activity of natural triterpene acids - meristotropic and macedonic acids - and their synthetic derivatives was studied. The activities of these compounds were assessed in relation to influenza viruses A and B in MDCK cell cultures and in a model of lethal influenza pneumonia in mice. The greatest levels of in vitro activity were seen with meristotropic acid and the methyl ester of macedonic acid. Acorrelation was noted between the activities of these compounds against influenza A and B viruses. At the same time, some of the study compounds lacked in vitro activity but had protective properties in vivo, which indicates that these have some other mechanism of action not associated with actions on virus-specific targets. These results allow derivatives of meristotropic and macedonic acids to be regarded as new effective antiviral agents.

Despite advances in the creation of vaccines and chemotherapeutic agents, influenza epidemics continue on a massive scale. Thus, there is great relevance in seeking and developing new effective therapeutic agents. Some triterpene acids are known to have marked antiviral and immunomodulatory activity [1, 2]. The aim of the present work was to study the antiviral activity of triterpene compounds, with an oleana-11,13(18)-diene skeleton in in vitro experiments (in MDCK cell cultures) and in vivo (in a model of lethal influenza in white mice).

Experimental Chemical Section

The structures of the compounds are shown in Fig. 1. ¹H NMR spectra were recorded on a Bruker AM-300 spectrometer (300 and 75.47 MHz) in deuterated methanol or deuterochloroform. IR spectra were taken on a Specord UR-75 spectrophotometer in Vaseline. UV spectra were recorded on an SF-46 spectrophotometer in ethanol.

Fig. 1

Triterpene acid derivatives studied here.

Meristotropic acid (3b-hydroxy-22-oxooleana-11,13(18)-dien-29-oic) acid (I). This was prepared by ethanol extraction of the dry roots of Glycyrrhiza triphylla Fich et Mey, syn. Meristotropis xanthoides Vass. [3]. The structure of this acid has been established previously [4 – 6]. The melting temperature was 355 – 358 °C. The IR spectrum, ν_max , cm^–1, was: 3630 (-OH), 1700 (C = O), 1610 (HC-C<); the UV spectrum, (ethanol), λ_max , nm, was: 242, 250, 258 (lg > 4). The ¹H NMR spectrum, (CDCl₃), δ, ppm, was: 0.78 (3H, CH₃, s), 0.80 (3H, CH₃, s), 0.90 (3H, CH₃, s), 0.95 (3H,CH₃, s), 0.99 (3H, CH₃, s), 1.29 (3H, CH₃, s), 1.34 (3H, CH₃,s), 1.96 (1H, H-9 broad s), 2.22 (1H, H-19 eq, d, J 14.5 Hz), 2.41 (1H, H-21 eq, d, J 16.1 Hz), 2.74 (1H, H-21 ax, dd, J₁ 16 Hz, J₂ 2.4 Hz), 3.24 dd (1H, H-19 ax, J₁ 14.5 Hz, J₂ 2.4 Hz), 3.26 (1H, H-3, dd, J₁ 11 Hz, J₂ 5 Hz), 5.65 (1H, H-12, d, J 10.5 Hz), 6.36 (1H, H-11, dd, J₁ 10.5 Hz, J₂ 3 Hz).

Methyl ester of 3𝛃-hydroxy-22-oxooleana-11,13(18)-dien-29-oic acid (II). Meristotropic acid (I) (3 g) was methylated using an ether solution of diazomethane prepared from 3 g of nitrosomethylurea [7] and methyl ester (II) was recrystallized from ethanol. The melting temperature was 280 – 282 °C. The IR spectrum, ν_max , cm^–1, was: 3620 (-OH), 1731, 1715 (C = O), 1610 (HC = C<). The ¹H NMR spectrum, (CDCl₃), δ, ppm, was: 0.75 (3H, CH₃, s), 0.79 (3H, CH₃, s), 0.87 (3H, CH₃, s), 0.91 (3H, CH₃, s), 1.00 (3H,CH₃, s), 1.27 (3H, CH₃, s), 1.34 (3H, CH₃, s), 1.91 (1H, H-9, broad s), 2.18 (1H, H-19 eq, d, J 14.5 Hz), 2.37 (1H, H-21 eq, d, J 16.1 Hz), 2.75 (1H, H-21 ax, dd, J₁ 16 Hz, J₂ 2.4 Hz), 3.24 (1H, H-19 ax, dd, J₁ 14.5 Hz, J₂ 2.4 Hz), 3.26 (1H, H-3, dd, J₁ 11 Hz, J₂ 5 Hz), 3.55 (3H, CH₃, s) 5.66 (1H, H-12, d, J 10.5 Hz), 6.36 (1H, H-11, dd, J₁ 10.5 Hz, J₂ 3 Hz).

Methyl ester of 3,22-dioxooleana-11,13(18)-dien-29-oic acid (III). Methyl ester (II) (2 g) in 8 ml of pyridine was oxidized with Jones reagent (3 g of CrO₃ in 50 ml of pyridine) [7]. At the end of the reaction, 20 ml of ethanol and 150 ml of water were added. The reaction product was extracted with ethyl acetate. The ethyl acetate solution was washed three times with water (30 ml each time), dried, and evaporated; chromatography of the dry residue by flash chromatography yielded 1.4 g of dioxo compound III. The melting temperature was 218 – 220 °C. The IR spectrum, ν_max , cm ^–1, was: 1731, 1715 (C = O), 1610 (HC = C<). The ¹H NMR spectrum, (CDCl₃), δ, ppm, was: 0.77 (3H, CH₃, s), 0.87 (3H, CH₃, s), 1.01 (3H, CH₃, s), 1.04 (3H, CH₃, s), 1.10 (3H, CH₃, s), 1.27 (3H, CH₃, s), 1.34 (3H, CH₃, s), 1.80 (1H, H-16 ax, dd, J₁ 14 Hz, J₂ 2.9 Hz), 1.81 (1H, H-16 eq, broad s), 2.00 (1H, H-9, broad s), 2.18 (1H, H-19 eq, d, J 13.8 Hz), 2.36 (1H, H-21 eq, d, J 16 Hz), 2.47 (1H, H-2 eq, m), 2.56 (1H, H-2 ax, m), 2.74 (1H, H-21 ax, dd, J₁ 16 Hz, J₂ 2.9 Hz), 3.22 (1H, H-19 ax, dd, J₁ 14.5 Hz, J₂ 2.5 Hz), 3.52 (3H, CH₃, s), 5.63 (1H, H-12, d, J 10 Hz), 6.39 (1H, H-11, dd, J₁ 11 Hz, J₂ 3 Hz).

3,22-Dioxooleana-11,13(18)-dien-29-oic acid (IV). Compound III (1 g) was boiled in 100 ml of 1 % ethanolic KOH for 3 h. The reaction mix was diluted with four volumes of water, acidified to pH 2 with HCl, and filtered; the precipitate was washed with water and dried. The yield was 0.9 g. The melting temperature was 285 – 287 °C. The IR spectrum, ν_max , cm^{– 1}, was: 1731, 1715 (C = O), 1610 (HC = C<). The ¹H NMR spectrum, (CDCl₃), δ, ppm, was: 0.80 (3H, CH₃, s), 0.85 (3H, CH₃, s), 1.03 (3H, CH₃, s), 1.05 (3H, CH₃, s), 1.11 (3H, CH₃, s), 1.29 (3H, CH₃, s), 1.37 (3H, CH₃, s), 1.81 (1H, H-16 eq, d, J 3.6 Hz), 1.82 (1H, H-16 ax, dd, J₁ 7.3 Hz, J₂ 2.9 Hz), 1.99 (1H, H-9 broad s), 2.21 (1H, H-19 eq, d, J 15 Hz), 2.37 (1H, H-21 eq, d, J 16 Hz), 2.48 (1H, H-2 eq, m), 2 58 (1H, H-2 ax, m), 2.77 (1H, H-21 ax, d, J₁ 16 Hz, J₂ 2.2 Hz), 3.28 (1H, H-19 ax, dd, J₁ 14.5 Hz, J₂ 2.2 Hz), 5.63 (1H, H-12, d, J 10 Hz), 6.45 (1H, H-11, dd, J₁ 10.5 Hz, J₂ 3 Hz).

Methyl ester of 3𝛃,22𝛃-dihydroxyoleana-11,13(18)-dien-29-oic acid (V). Compound II (1 g) was dissolved in 50 ml of propan-2-ol and a solution of 1 g of NaBH₄ in 10 ml of propan-2-ol and 10 ml of water were added. The reaction mix was boiled for 4 h, cooled, diluted with water, and extracted with chloroform. The chloroform extract was washed with water, dried, chromatographed by the flash chromatography method, and eluted with chloroform. The yield of compound V was 0.4 g. The melting temperature was 263 – 265 °C (4). The IR spectrum, ν_max, cm^{– 1}, was: 3630, 1735 (C = O). The ¹H NMR spectrum, (CDCl₃), δ, ppm, was: 0.74 (3H, CH₃, s), 0.80 (3H, CH₃, s), 0.89 (3H, CH₃, s), 0.91 (3H, CH₃, s), 1.01 (3H, CH₃, s), 1.06 (3H, CH₃, s), 1.25 (3H, CH₃, s), 1.92 (1H, H-9 broad s), 2.26 (1H, H-19 eq, dd, J₁ 11.6 Hz, J₂ 3 Hz), 3.16 (1H, H-19 ax, d, J 14.5 Hz), 3.26 (1H, H-3, dd, J₁ 11 Hz, J₂ 5 Hz), 3.41 (1H, H-22 ax, dd, J₁ 12 Hz, J₂ 4 Hz), 3.60 (3H, CH₃, s), 5.64 (1H, H-12, d, J 11 Hz), 6.36 (1H, H-11, dd, J₁ 11.1 Hz, J₂ 3 Hz).

Methyl ester of 3𝛃,21𝛂-dihydroxyoleana-11,13(18)-dien-29-oic acid (VII). Macedonic (3β,21α-dihydroxyoleana-11,13(18)-dien-29-oic) acid (VI) was extracted from Glycyrrhiza macedonica Boiss et Orph. roots with methanol as described in [1]. The melting temperature was 340 – 343 °C. M⁺ (m/z) 470. The UV spectrum, (ethanol), ν, cm ^{– 1}, was: 239, 248, 259 (lg ε 4.4, 4.5, 4.3). Acid VI was methylated using an ether solution of diazomethane [7] by the same method used for meristotropic acid. The resulting methyl ester VII had a melting temperature of 252 – 254 °C [4, 5]. The IR spectrum, ν_max, cm^{– 1}, was: 3620, 1730 (C = O). The ¹H NMR spectrum, (CDCl₃), δ, ppm, was: 0.69 (3H, CH₃, s), 0.77 (3H, CH₃, s), 0.87 (3H, CH₃, s), 0.89 (3H, CH₃, s), 0.98 (3H, CH₃, s), 1.11 (3H, CH₃, s), 1.39 (3H, CH₃, s), 3.24 (2H, m), 3.59 (3H, CH₃, s), 5.60 (1H, H-12, d, J 10.5 Hz), 6.46 (1H, H-11, dd, J₁ 10.5 Hz, J₂ 3 Hz).

Methyl ester of 3,21-dioxooleana-11,13(18)-dien-29-oic acid (VIII). Methyl ester VII (0.53 g) was oxidized with Jones reagent as described for the oxidation of compound II. Oxidation products were separated by flash chromatography and eluted with a mixture of hexane and ethyl acetate (1:1). The yield of product VIII was 0.45 g. The melting temperature was 242 – 242 °C. The IR spectrum, ν_max, cm^{– 1}, was: 1730, 1710 (C = O). The ¹H NMR spectrum, (CDCl₃), δ, ppm, was: 0.76 (3H, CH₃, s), 0.91 (3H, CH₃, s), 1.03 (3H, CH₃, s), 1.06 (3H, CH₃, s), 1.12 (3H, CH₃, s), 1.15 (3H, CH₃, s), 1.42 (3H, CH₃, s), 2.02 (1H, H-9, broad s), 3.18 (1H, H-19 ax, d, J 14.5 Hz), 3.62 (3H, CH₃, s), 5.60 (1H, H-12, d, J 10.2 Hz), 6.53 (1H, H-11, dd, J₁ 10.2 Hz, J₂ 3 Hz).

Diacetoxy 3𝛃,21𝛂-dihydroxyoleana-11,13(18)-dien-29-oic acid (IX). Macedonic acid (VI) (1 g), 10 ml of pyridine, and 12 ml of acetic anhydride were boiled for 4 h. The reaction mix was diluted with three volumes of water, the precipitate was collected by filtration, washed with water, and recrystallized from ethyl acetate. The melting temperature was 290 – 292 °C. The IR spectrum, ν_max, cm^{– 1}, was: 1730, 1710, 1620, 1255. The ¹H NMR spectrum, (CDCl₃), δ, ppm, was: 0.72 (3H, CH₃, s), 0.86 (3H, CH₃, s), 0.88 (3H, CH₃, s), 0.91 (3H, CH₃, s), 0.92 (3H, CH₃, s), 1.21 (3H, CH₃, s), 1.27 (3H, CH₃, s), 1.92 (1H, H-9, broad s), 1.96 (1H, H-19 ax, dd, J₁ 15.2 Hz, J₂ 3 Hz), 2.07 (3H, CH₃, s), 2.08 (3H, CH₃, s), 3.12 (1H, H-19 eq, d, J 15.2 Hz), 4.53 (1H, H-3, dd, J₁ 10.2 Hz, J₂ 6.5 Hz), 5.06 (1H, H-21, dd, J₁ 12.3 Hz, J₂ 4.7 Hz), 5.58 (1H, H-12, d, J 11 Hz), 6.38 (1H, H-11, dd, J₁ 11 Hz, J₂ 3 Hz).

Glycyrrhetic (3𝛃-hydroxy-11-oxooleana-12-en-30-oic) acid (X). Glycyrrhetic acid (X) was extracted from Glycyrrhiza glabra L. roots as described in [6]. The melting temperature was 293 – 296 °C. The ¹H NMR spectrum, (CD₃OD), δ, ppm, was: 0.82 (3H, CH₃, s), 0.85 (3H, CH₃, s), 1.01 (3H, CH₃, s), 1.16 (6H, CH₃, s), 1.19 (3H, CH₃, s), 1.44 (3H, CH₃, s), 2.47 (1H, H-9, s), 2.74 (1H, H-18 dd, J₁ 13.8 Hz, J₂ 3.6 Hz), 3.19 (1H, H-13 dd, J₁ 11.6 Hz, J₂ 5 Hz), 5.6 (1H, H-12, s).

Methyl ester of glycyrrhetic acid (XI). Glycyrrhetic acid (X) was methylated using an ether solution of diazomethane [7]. The methyl ester (XI) had a melting temperature of 248 – 251 °C. The ¹H NMR spectrum, (CDCl₃), δ, ppm, was: 0.82 (3H, CH₃, s), 0.83 (3H, CH₃, s), 1.01 (3H, CH₃, s), 1.13 (3H, CH₃, s), 1.14 (3H, CH₃, s), 1.16 (3H, CH₃, s), 1.37 (3H, CH₃, s), 2.35 (1H, H-9, s), 2.80 (1H, H-18, dd J₁ 13 Hz, J₂ 3.3 Hz), 3.24 (1H, H-3, dd, J₁ 10.2 Hz, J₂ 5.8 Hz), 3.70 (3H, s), 5.67 (1H, H-12, s).

Experimental Biological Section

The following influenza viruses, from the virus collection of the Science Research Institute of Influenza, Ministry of Health of the Russian Federation, were used: A/Puerto Rico/8/34 (H1N1), A/Aichi/2/68 (H3N2), and B/Lee/40.

Influenza viruses were passaged in the allantoic cavities of chick embryos aged 10 – 12 days for 48 h at 36 °C (influenza A viruses) or for 72 h at 34 °C (influenza B virus).

Assessment of cytotoxicity and antiviral activity of compounds in cell cultures. The toxicity of compounds was evaluated by incubating serial two-fold dilutions with MDCK cells for 48 h at 37 °C in a 5 % CO₂ atmosphere. After incubation, cells were washed for two 5-min periods with phosphate-saline buffer, after which the numbers of live cells in wells were assayed using the micro tetrazolium test (MTT) [8]: 100 μl of a solution (5 mg/ml) of 3-(4,5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide (ICN Biochemicals Inc., Aurora, Ohio) in physiological saline was added to each well, cells were incubated at 37 °C in a 5 % CO₂ atmosphere for 2 h, and washed with phosphate-salt buffer for 5 min. Precipitates were dissolved in DMSO (100 μl/well), after which the optical density in each well was measured using a Victor 1420 multifunctional reader (Perkin Elmer, Finland) at a wavelength of 535 nm. Test results were used to determine the 50 % cytotoxic dose (CTD₅₀) for each compound, i.e., the concentration causing death of 50 % of cells in cultures.

The antiviral activity of compounds was assayed by adding compounds at the appropriate concentrations to wells containing cell monolayers and incubating for 1 h at 37 °C; cultures were then infected with ten-fold (10 – 1 to 10 – 5) serial dilutions of virus. After infection, cells were incubated for 37 °C in a 5 % CO₂ atmosphere for 48 h (influenza A viruses) or for 72 h (influenza B virus). Controls consisted of cell cultures infected with virus (positive controls) and intact cell cultures treated with nutrient medium in place of compounds (negative controls). After incubation, 50 μl of culture fluid was transferred to the corresponding round-bottom wells of a 96-well plate (Medpolimer, St. Petersburg) and 50 μl of 1 % suspension of chick erythrocytes in physiological saline was added to each well. Plates were incubated for 1 h at room temperature and the presence of virus was assessed in terms of erythrocyte agglutination in wells. The hemagglutination titer of the virus was expressed as the base 10 log of the dilution of starting material giving positive hemagglutination reactions in wells. Titers were given as log 50 % infective doses (lg ID₅₀). Test results for each compound were used to determine the 50 % effective dose (ED₅₀), i.e., the concentration halving virus replication, and the chemotherapeutic index (CTI, the ratio of CTD₅₀ to ED₅₀). Antiviral activity was evaluated as the CTI.

Experimental influenza infection. White mongrel mice (females) weighing 14 – 16 g were obtained from the Rappolovo supplier (Leningrad Region) and were kept on a standard diet in controlled animal-house conditions at the Science Research Institute of Influenza, Russian Academy of Medical Sciences. Animals were grouped randomly. Animals were monitored for two weeks before the study. Before experiments, five mice were infected with 0.05 ml of allantoic fluid containing influenza virus A/Aichi/2/68 (H3N2) (5 × 107 ID₅₀/ml). Lungs were harvested three days after infection and homogenized in 10 volumes of sterile physiological saline, after which virus activity in homogenates was assayed in separate experiments by titration for lethality in animals. Virus titers were calculated as described by Reed and Muench [9].

Doses killing 50 % of animals, i.e., 50 % lethal doses (LD₅₀), were determined for each agent. Animals received study compounds i.p. in a volume of 0.2 ml using a therapeutic-prophylactic regime: 24 h and 1 h before infection and 24, 48, and 72 h after infection at a dose of 1/3 LD₅₀. Animals of the control group received physiological phosphate buffer. The reference agent was remantadine (Aldrich Chem. Co.) at a dose of 50 mg/kg. Remantadine was given i.p. using the same regime as for test compounds. The negative control consisted of intact animals kept in the same conditions as the experimental animals.

Animals received virus intranasally under light anesthesia at a dose of 1 LD₅₀. Each group consisted of 15 mice. Animals were observed for 14 days, i.e., the period during which experimental influenza caused death. Lethality among the animals in the control and experimental groups was assessed daily. Lethality data were used to calculate the proportions of animals dying in each group (ratio of animals dead by 14 days to total number of animals infected); the mean duration of life was measured, along with the index of protection (IP, ratio of the difference in the proportions dying in the control and experiments groups to the proportion dying in the control group). The antiviral activity of compounds was assessed in terms of the IP value.

Results and Discussion

The results are summarized in Table 1, 2.

Table 1 Antiviral Properties of Triterpene Acids against Influenza Viruses in Vitro.

Table 2 In Vivo Antiviral Properties of Triterpene Acids against Influenza Viruses.

Results on the activity of the compounds against influenza virus A (H1N1) show that the highest levels of activity were seen with meristotropic acid (I) and the methyl ester of macedonic acid (VII). The activities of these compounds were greater than that of the reference compound remantadine. The influenza A virus A/PR/8/34 (H1N1) used ere is known to be resistant to remantadine, such that the high levels of activity seen with compounds I and VII indicate that they have potential for further development as agents for protection against influenza. The other compounds had no marked virus-inhibiting properties. There was a correlation between the activities of these compounds against influenza A and B viruses.

Overall, it can be noted that virtually all the modifications to the natural compounds studied here led to loss of antiviral activity (compound I and its derivatives II-V, X, and XI). The exception was the presence of activity with the methyl ester of macedonic acid (VII) when it was absent from the acid itself (VI). Analysis of ED₅₀ values showed that compounds II and X had antiviral activity, though their high toxicity led to sharp reductions in CTI.

At the same time, some of the triterpene acids studied here (IX, X, XI) had protective properties in vivo. These properties were apparent as decreases in specific death and increases in the duration of life as compared with the control groups. The greatest activity was seen with compounds VII and X (index of protection 37.5 %). Nonetheless, none of the study compounds was more active than reference agent remantadine (index of protection 75 %). The only agent producing a significant increase in the duration of life compared with controls was remantadine. The combination of overall protective activity in vivo with direct antiviral activity seen in the present study with compounds I and VII provides evidence that they have potential in the prophylaxis and/or treatment of influenza infections in humans. The current task in this direction consists of testing their activity against contemporary epidemic viruses, including the pandemic influenza strain A(H1N1)pdm09. The results obtained here point to the potential for further development of meristotropic and macedonic acids and their derivatives for the creation of new effective anti-influenza agents.