Antibacterial activity of some chemical constituents from Trichilia prieuriana (Meliaceae)

2.1 Isolation and identification of compounds

Twenty-two (22) compounds were isolated through the bio-guided fractionation of the hydro-ethanolic extracts from different parts of T. prieuriana by usual chromatographic techniques. Their structures were established by spectroscopic methods and by comparison to data of related compounds described in the literature, while the molecular structures of compounds 1 and 2 were confirmed by single-crystal X-ray (SC-XRD) diffraction analyses.

Compound 1 was isolated as colourless crystals from the n-hexane-EtOAc (3:2) fraction of the root wood of T. Prieuriana. It gave a red colour in the Liebermann-Burchard reaction suggesting its steroidic character. Its molecular formula, C₂₁H₃₄O₄ (five degrees of unsaturation) was deduced from its HR-ESI-MS spectrum (positive mode), where sodium adduct ion peak was observed at m/z = 373.2395 [M + Na]⁺ (calcd. 373.2349 for C₂₁H₃₄O₄Na). Its IR spectrum exhibited vibration bands at 3324 and 1735 cm⁻¹ due to the hydroxyl and five-membered ring ketone, respectively. The ¹H NMR data (Table S1) showed resonance due to one methyl triplet at δ = 1.02 ppm (3H, t, J = 12 Hz), two methyl singlets at δ = 0.70 ppm (3H, s) and δ _H = 1.26 ppm (3H, s), and three oxymethine protons at δ = 4.14 (H, brd) 3.53 (H, brd) and 3.84 ppm (H, brd). The ¹³C NMR spectrum showed resonances signals which were sorted into three methyl groups (CH₃) at δ = 13.5; 13.7 and 17.4 ppm; seven (07) methylene groups (CH₂) all sp ³-hybridized at δ = 17.2; 20.2; 18.3; 32.5; 38.2; 38.6 and 43.2 ppm; eight (08) methine groups (CH) all sp ³-hybridized at δ = 34.1; 50.5; 50.6; 56.8 and 65.2 ppm among which three (03) oxymethine groups (CH–OH) at δ = 71.8, 72,3 and 76.7 ppm and one keto carbonyl group at δ = 219.5 ppm. The above structural features together with correlations from ¹H-¹H COSY, HSQC and HMBC experiments suggested that compound 1 is 2β,3β,4β-trihydroxypregnan-16-one [12] (see Figure 1).

Figure 1:

Chemical structures of compounds 1–22 isolated from Trichilia prieuriana.

Compound 2 was isolated as colourless crystals from the n-hexane-EtOAc (3:2) fraction of the stem bark of T. Prieuriana. It gave a red colour in the Liebermann–Burchard reaction and a pink colour in the Ehrlich test, suggesting its limonoïdic nature. Its molecular formula, C₃₈H₅₀O₁₆ (14° of unsaturation) was deduced from its HR-ESI-MS spectrum (positive mode). The ion peak of the sodium adduct was observed at m/z = 785.3004 [M + Na]⁺ (calcd. 785.2991 for C₃₈H₅₀O₁₆Na). Its IR spectrum showed maximum absorption bands at 3067 and 1740 cm⁻¹ characteristic of hydroxyl and carbonyl groups, respectively. Based on the HR-ESI-MS, ¹H NMR and ¹³C NMR data (Table S2), X-ray crystallographic data (see below) and by spectral comparison with previous reported values, compound 2 was elucidated to be prieurianin [13] (see Figure 1).

Confirmation of the molecular structures and the absolute stereochemistry of compounds 1 and 2 were obtained by SC-XRD analysis (see below).

The 20 other compounds (3–22) were identified as: flindissone (3) [14], deoxyflindissone (4) [15], picraquassin E (5) [16], ursolic acid (6) [17], 3β-acetoxy-11α-hydroxyurs-12-en (7) [18], 3β-acetoxy-urs-12-en-11-one (8) [19], 3β-acetoxy-β-amyrin (9) [20], friedelin-3-ol (10) [21], 3-oxo, friedelin (11) [22], 3oxo, fridelin-28-ol (12) [23], oleanolic acid (13) [24] hederagenin (14) [25], stigmasterol (15) [26], β-sitosterol (16) [26] and their glycosylated derivatives β-sitosterol-3-O-β-glucopyranoside (17) [27] and stigmasterol-3-O-β-glucopyranoside (18) [27], erythrodiol (19) [28], scopoletin (20) [29], 4-hydroxy-3,5-dimethoxybenzoic acid (21) [30] shikimic acid (22) [31] (see Figure 1). The NMR data of compounds 1 and 2 are given in the Supporting Information available online.

2.2 X-ray crystal structure determinations

The crystallization of compounds 1 and 2 from MeOH and a mixture of acetone-CH₂Cl₂-MeOH (0.5:1:1), respectively, provided colourless crystals. The successful refinement of the Flack x parameters (see Section 3) allowed the determination of the absolute structures of 1 and 2 as they are depicted in Figures 1 and 2.

Figure 2:

Ortep drawing of the molecular structures of compounds 1 and 2 in the crystal. Only the asymmetric carbon atoms were labelled. Displacement ellipsoids were drawn at the 50% probability level.

Compound 1 shows 10 asymmetric carbon atoms which turned out to be C3 S; C4 S; C7 R; C8 R; C9 S; C10 S; C12 R; C13 S; C16 S and C17 R (Figure 2a shows atom labels). In compound 2, the chirality of the twelve stereo-centres could be determined as follows: C3 is S; C4 R; C5 R; C6 S; C7 R; C8 R; C9 R; C10 R, C11 S; C14 S; C29 R and C30, R (Figure 2b shows atom labels).

2.3 Antibacterial potential of extracts, fractions and compounds

The antibacterial activity of extracts, fractions and compounds (1–22) was evaluated against nine pathogenic bacteria strains using the micro broth dilution method [32] The nine test bacterial strains were Staphylococcus aureus CHU, Shigella flexneri NR 518, Klebsiella pneumoniae ATCC 700603, Escherichia coli ATCC 25922, Salmonella enterica anatum, S. aureus BAA 917, S. aureus ATCC 43300, E. coli CHU and Streptococcus pneumoniae ATCC 49619. The tested crude extracts, fractions and compounds exhibited varied levels of antibacterial activities against the test bacteria as shown in Table S3. From the results, the MIC values of crude extracts ranged from 31.25 to 500 µg mL⁻¹, with the root wood extract being the most potent. All tested fractions exhibited moderate activity on the test bacterial strains with MIC values of 250 µg mL⁻¹.

The isolated compounds showed the best activity profile, with flindissone (3), ursolic acid (6) and 3-oxo, fridelin-28-ol (12) being the most potent with a broad spectrum activity (MIC varying from 4.09 to 71.8 µm). The Gram-negative E. coli ATCC 25922 appeared to be the most susceptible strain (MIC varying from 4.09 to 41.66 µm). The antibacterial activity displayed by the ethanolic crude extract from the root wood in this study was greater than that reported by Kuglerova et al. [33], who reported a weak antibacterial potency for the bark extract of T. prieuriana against Enterococcus faecalis and S. aureus (MIC of 512 and 256 µg mL⁻¹ respectively). Data from the present study coupled with those previously reported by other researchers on plants from Meliaceae family such as Trichilia emetica [34], also indicated that limonoids, prieurianin (1), flindissone (3), deoxyflindissone (4) and picraquassin E (5) isolated from root wood extract could be responsible for the exhibited antibacterial activity. Indeed, the evaluated compounds showed inhibitory potential against three of the reference strains used. Isolated phytosterols such as β-sitosterol have been proven by many research groups to possess good antibacterial potential against several pathogenic bacteria including S. aureus [35, 36]. Extracts and compounds inhibited mostly Gram-negative compared to Gram-positive bacteria. This is not in accordance with the findings of other researchers who demonstrated that Gram-negative bacteria were more resistant to plant-based organic extracts because their hydrophilic cell wall structure is essentially constituted of lipopolysaccharides that block the penetration of hydrophobic oil thereby preventing the accumulation of organic extract in their cell membrane. However, the enhanced susceptibility of Gram-negative bacteria noticed in this work is of particular interest, given their significant impact in the global antimicrobial resistance outbreak. Indeed, in Gram-negative bacteria, intrinsic resistance to currently available antibiotics is mainly due to overexpressed efflux pumps which are constitutively present and also the presence of protective outer membrane that prevent drugs to reach their metabolic targets in effective doses.

3.1 General experimental procedures

IR spectra were recorded on a Bruker Fourier transform/infrared (ATR) spectrophotometer. Mass spectra (ESI-MS) were obtained with a Thermo-Finnigan LCQ DECA mass spectrometer and HR-ESI-MS spectra were measured with a FTHRMS-Orbitrap (Thermo-Finnigan) mass spectrometer. 1D and 2D-NMR spectra were recorded in deuterated solvents on either Bruker ARX 500 or AVANCE DMX 600 NMR spectrometers (proton at 600 MHz and carbon 13C at 150 MHz). All chemical shifts (δ) were measured in parts per million (ppm) using a residual solvent signal as secondary reference relatively to tetramethylsilane (TMS) as internal standard, while coupling constants (J) are given in Hz. Solvents were distilled prior to use. Analytical grade solvents were used for LC/MS. Column chromatography (CC) was performed using Merck MN silica gel 60 M (0.04–0.063 nm) and thin layer chromatography (TLC) was performed on aluminium silica gel 60 F254 (Merck) precoated plates (0.2 mm layer thickness). Spots were visualized on TLC either by UV lamp (254 and 365 nm) or by heating after spraying with 20% H₂SO₄ (v/v) solution. Different mixtures of n-hexane, EtOAc and MeOH were used as eluting solvents.

3.2 Plant material

Leaves, root wood, stem bark and root bark of T. prieuriana were collected in August 2016 at Mindourou (GPS coordinates: Latitude 3.98250 North and Longitude 14.89501 East), East region, Cameroon. Plant material was identified by Dr. Nole, plant taxonomist at the Institute of Medical Research and Medicinal Plants Studies (IMPM), Cameroon. Voucher specimens are preserved in the National Herbarium of Cameroon under the reference number 66990/HNC.

3.3 Extraction, fractionation and isolation

Dried ground leaves (2.3 kg), root wood (1.9 kg), stem bark (1.1 kg) and root bark (1.15 kg) of T. prieuriana were extracted three times with the mixture of ethanol–water (7:3) at room temperature for 48 h each time. The filtrate from each part was combined and evaporated to dryness to give 208 g of a green residue (leaves); 150 g, 387 and 44 g of a brown residues for root wood, stem bark and root bark respectively.

One hundred and 45 g of the hydro-ethanolic crude extract obtained from the root wood was subjected to flash chromatography over silica gel using a mixture of n-hexane and AcOEt of increasing polarity to afford four main fractions labelled F1 (20 g; n-hexane- EtOAc 4:1), F2 (10 g; n-hexane-EtOAc 1:1), F3 (15 g; pure EtOAc), and F4 (25 g; MeOH). Fraction F1 (20 g) was further chromatographed on a silica gel column and eluted with a mixture of n-hexane-EtOAc of increasing polarity. A total of 200 fractions, each of 175 mL, were collected and combined based on their TLC profiles to afford seven main subfractions [S1 (1–30), S2 (31–60), S3 (61–86), S4 (87–115), S5 (116–148), S6 (149–180) and S7 (181–200)]. Series S1, S2, S3, S4 and S5, after precipitation at room temperature, followed by filtration and washing with a mixture of n-hexane-EtOAc (1:1) yielded compounds 2 (30.09 mg), 3 (2 mg), 5 (30.68 mg), 15 (2056 mg) and 16 (25.23 mg), respectively. Fraction F2 (10 g) obtained by elution with n-hexane- EtOAc (1:1), was further purified over a silica gel column using a gradient of n-hexane-EtOAc to give a total 160 fractions of 100 mL each, which were further combined based on TLC profiles to afford nine main series [S1′ (1–15), S2′ (16–35), S3′ (35–48), S4′ (49–60), S5′ (60–80), S6′ (80–105), S7′ (106–120), S8′ (106–137), S9′ (138–160)]. Series S4′ and S6′ crystallized and after filtration, yielded compounds 1 (2.56 mg) and 4 (8.56 mg). Then S7′ was also crystallized at room temperature, and after filtration compounds 11 (20.67 mg) and 13 (25.33 mg) were obtained.

The crude extract from root wood (39.3 g) was chromatographed over silica gel column, eluting with n-hexane-EtOAc mixtures with increasing proportion of EtOAc. 200 mL fractions were collected, evaporated and pooled based on their profiles on the analytical TLC plate to give 15 main series [S1′ (1–11), S2′(12–39), S3′ (40–54), S4′ (55–106), S5′ (107–118), S6′ (119–156), S7′ (157–181), S8′ (182–211), S9′ (212–226), S10′ (227–314), S11′(315–348), S12′ (349–392), S13′ (393–433), S14′ (434–410), S15′ (511–539)]. Fractions 157 to 181, once combined and left at room temperature, precipitated to give, after filtration and washing, compound 15 (10.56 mg) as colourless powder. Similarly, fractions 315 to 348 eluted with n-hexane-EtOAc (1:1) afforded a pure compound 16 (10.76 mg) after evaporation of the solvent. Fractions 349 to 360, when combined and left at room temperature, precipitated in the form of colourless needles, to give compound 21 (8.89 mg), after filtration and washing with methanol. With a similar procedure, fractions 379–388 led to compound 22 (24.80 mg).

Crude extracts (376.44 g) from stem bark were subjected to flash chromatography over silica gel using n-hexane-EtOAc mixtures with increasing proportion of EtOAc gave five main fractions labelled F1 (58 g; n-hexane-EtOAc, 3 :1), F2 (11 g; n-hexane-EtOAc, 1:1), F3 (2 g n-hexane-EtOAc 1 :3), F4 (3 g; pure EtOAc), and F5 (23 g; MeOH).

Of 58 g obtained from fraction F1, 54 g were chromatographed over silica gel column (63–200 nm mesh particle size), eluting with n-hexane-EtOAc mixtures of increasing polarity. 240 sub-fractions of 175 mL each were collected and pooled into seven main sub-fractions (S1 [1–17], S2 [18–62], S3 [63–130], S4 [131–162], S5 [163–182], S6 [183–220] and S7 [221–240]), according to their TLC profiles. After precipitation, filtration and washing with methanol, sub-fractions 92–101 led to compound 7 (12.86 mg). The S6 series obtained by evaporation of the solvents from sub-fractions 183 to 220, were further subjected to chromatography over silica gel column and eluted iso cratically with the mixture of n-hexane-EtOAc (3:1). From the 40 sub-fractions collected, sub-fractions 20–28 yielded compound 6 (10.65 mg). Fraction F2 (11 g) obtained from the flash chromatography was chromatographed on a silica gel column (particle size of 63–200 nm mesh) and eluted with the n-hexane-EtOAc of increasing polarity. 245 sub-fractions of 200 mL each were collected and pooled into 10 sub-fractions (S1 [1–10], S2 [11–41], S3 [42–80], S4 [81–103], S5 [104–136], S6 [137–160], S7 [161–181], S8 [182–202], S9 [203–221] and S10 [222–245]), based on their analytical TLC profiles. Fractions 137–160 were combined after evaporation of the solvent to yield at room temperature, a green precipitate, which was further filtrated and washed with methanol to give compound 19 (10.04 mg). Sub-fractions 161–181 were left at room temperature for one day, and on crystallization with methanol gave compound 20 (12.04 mg) in the form of purple crystals.

Crude extract from leaves (182 g) were fractionated by flash chromatography over silica gel, eluting with n-hexane-EtOAc mixture of increasing polarity to give five fractions labelled F1 (25 g; n-hexane-EtOAc, 3:1), F2 (10 g; n-hexane-EtOAc, 1:1), F3 (9 g n-hexane-EtOAc 1:3), F4 (5 g; pure EtOAc), and F5 (15 g; MeOH).

Fraction F1 (20 g) was further chromatographed over silica gel column (63–200 nm mesh) and eluted with n-hexane-EtOAc mixtures of increasing polarity. 230 sub-fractions of 200 mL each were collected and grouped into 11 sub-fractions (S1 [1–10], S2 [11–55], S3 [56–86], S4 [87–116], S5 [117–134], S6 [135–155], S7 [156–171], S8 [172–182], S9 [183–204], S10 [205–211] and S11 [212–230]). From sub-fractions 82–86, when combined and left at room temperature, compound 3 (18.33 mg) precipitates as a colourless powder. Sub-fractions 183–204, when left at room temperature for a day, crystallized and yielded after filtration, a whitish-coloured powder, compound 6 (10.45 mg).

Fractions F2 (11 g) and F3 (2 g) were combined based on analytical TLC profiles. The resulting fraction was chromatographed over silica gel column (63–200 nm mesh) and eluted with the n-hexane-EtOAc mixture with increasing polarity of EtOAc. 260 sub-fractions of 200 mL each were collected and grouped into 9 main sub-fractions (S1 [1–31], S2 [32–55], S3 [56–86], S4 [87–117], S5 [118–141], S6 [142–168], S7 [169–200], S8 [201–240] and S9 [241–260]). Series S3 crystallized at room temperature to give compound 20 (10.05 mg) after filtration and washing with methanol. Similarly, series S5 after evaporation of the solvent gave a powder, which, when washed with ethyl acetate, resulted in compound 9 (14.42 mg). Series S7, (169–200) when left at room temperature for one day, precipitated as colourless powder to give compound 17 (8.67 mg). After evaporation of the solvent, from series S8, a precipitate was formed and it yielded compound 18 (9.48 mg) after washing with ethyl acetate. Sub-fractions 241–260 were left at room temperature for one day, and precipitated to give compound 13 (10.22 mg) as colourless powder.

3.4 Spectroscopic data of compounds 1 and 2

4.2-2β,3β,4β -trihydroxypregnan-16-one (1): Colourless crystals, m.p. 300–302 °C. – UV (MeOH): λ _max (log ɛ _max) = 218 nm (0.68), 230 (1.36). –  [ α ] D 20  = −95.2° (c = 0.05, MeOH). – IR: υ _max = 3324 (–OH), 2980–2840 (–CH₃, –CH₂), 1735 cm⁻¹ (C=O). – HR-ESI-MS: m/z = 373.2395 (calcd. 373.2349 for C₂₁H₃₄O₄Na, [M + Na]⁺). – ¹H NMR (500 MHz, CD₃OD) and ¹³C NMR (125 MHz, CD₃OD): data: see Table S1, Supplementary material.

Prieurianin (2): Colourless crystals, m.p. 230–232 °C. – [ α ] D 20  = −13.3° (c = 1.22, CH₃Cl). – IR: υ _max = 3067, 1740 cm^–1 – HR-ESI-MS: m/z = 785.3004 (calcd. 785.2991 for C₃₈H₅₀O₁₆Na, [M + Na]⁺). – ¹H NMR (500 MHz) and ¹³C NMR (125 MHz) data in CDCl₃: (see Table S2, Supplementary material);

3.5 X-ray crystal structure determinations

Single crystals of 1 and 2 were examined on a Rigaku Supernova diffractometer using CuKα radiation (λ = 1.54184 Å). The crystals were kept at T = 100.0(1) K during data collection. Using Olex2 [37], the structures were solved with Shelxt [38] using Intrinsic Phasing and refined with Shelxl [39] using least-squares minimisation.

Crystallographic data of 1 (C₂₁H₃₄O₄, M = 350.48 g mol⁻¹): orthorhombic, space group P2₁2₁2₁ (no. 19), a = 7.69568(7) Å, b = 13.16355(15) Å, c = 18.2969(2) Å, V = 1853.52(3) Å³, Z = 4, μ(CuKα) = 0.675 mm⁻¹, D _calc = 1.256 g cm⁻³. 30994 measured reflections (8.3 ≤ 2θ ≤ 152.7°) 3869 of which were unique (R _int = 0.0280, R _σ  = 0.0141) which were used in all further calculations. All hydrogen atoms were refined isotropically. The final R ₁ was 0.0302 for 3770 reflections with I > 2σ(I) and 362 refined parameters, while wR ₂ was 0.0890 for all data, Flack x = −0.06(5), Δρ _fin (max/min) = 0.23/−0.14 e Å⁻³.

Crystal data of 2 (C₃₈H₅₀O₁₆, M = 2.78 g mol⁻¹): orthorhombic, space group P2₁2₁2₁ (no. 19), a = 11.01447(10) Å, b = 11.78408(16) Å, c = 29.1402(3) Å, V = 3782.26(7) Å³, Z = 4, μ(CuKα) = 0.879 mm⁻¹, D _calc = 1.340 g cm⁻³, 66,605 measured reflections (6.1 ≤ 2θ ≤ 153.1°), 7886 of which unique (R _int = 0.0266, R _σ  = 0.0126) which were used in all further calculations. Hydrogen atoms were refined using a riding model except donor hydrogen atoms, which were refined isotropically. The final R ₁ was 0.0392 for 7813 reflections with I > 2σ(I) and 504 refined parameters, while wR ₂ was 0.1163 for all data, Flack x = 0.04(3), Δρ _fin (max/min) = 0.32/−0.19 e Å⁻³.

CCDC 2063671 and 2063672 contain the supplementary crystallographic data for 1 and 2. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/conts/retrieving.html.

3.6 Biological assays

3.7 Statistical analysis

Antibacterial experiments were carried out in triplicate. Data were expressed as mean ± standard deviation from triplicate values. P values of 0.05 or less were considered statistically significant.