Antimicrobial Activity of Cinnamomum impressicostatum and C. pubescens and Bioassay-Guided Isolation of Bioactive (E)-Methyl Cinnamate

MATERIALS AND METHODS

General experimental procedures: Melting points were measured on a Kofler hot stage apparatus and are uncorrected. The IR spectra were recorded using KBr discs on Perkin Elmer FTIR spectrophotometer model 1275X. The UV spectrum was recorded on a Shimadzu UV 160A spectrophotometer in MeOH. The ¹H-NMR and ¹³C-NMR spectra were obtained on a JOEL FTNMR 400 MHz spectrometer operating at 400 and 100 MHz, respectively. Chemical shifts are shown in δ values (ppm) with tetramethylsilane as an internal standard. Qualitative analysis by GC and GC/MS were performed on Shimadzu GC and Hewlett-Packard GCMSD apparatus with SE-30 capillary column.

Test organisms: Nine test organisms were used in the study. The bacterial strains used were Straphylococcus aureus, S. epidermidis, Bacillus cereus, Pseudomonas aeruginosa, Burkholderia cepacia. The four fungal strains used were yeast-like fungi Candida albicans, C. lipolytica, Saccharomyces cerevisae and Microsporum canis. M. canis was obtained from the American Type Culture Collection (ATCC) while the rest of the strains were obtained from Northern Regional Research Laboratories (NRRL) and College of Pharmacy, University of Iowa. The strains were cultured at the Animal Tissue Culture Laboratory, Universiti Putra Malaysia. The bacteria and fungi were maintained on nutrient agar and potato dextrose agar slants, respectively and stored at 4°C.

Disc diffusion assay: The antimicrobial activities of the samples were determined by the disc diffusion method (Bauer et al., 1966). Antibiotic discs of 6 mm diameter were prepared by impregnating them with 5 μL of each sample. The disc were then dried under sterile condition and then placed on the agar plate previously inoculated with the suspension of each microorganism to be tested. Streptomycin and nystatin were used as standard and methanol as negative control. The agar plates were then incubated for 24 h at 30°C and the antimicrobial activities were recorded my measuring the zone of inhibition around each disc. The experiments were carried out in triplicates and averaged.

Plant material and isolation of compounds: The barks of C. impressicostatum and C. pubescens were collected from Bukit Larut, Taiping and Gunung Berembun, Cameron Highlands, Perak in 1999, respectively. Voucher specimen were deposited at the Herbarium, Forest Research Institute Malaysia, Kepong, Malaysia.

Extraction of the plant samples: The air-dried plant samples were extracted exhaustively and separately with hexane, chloroform and methanol each over a period of two weeks. The extracts were then concentrated under reduced pressure and stored in freezer until required for analysis. The hexane, chloroform and methanol extraction of bark of C. impressicostatum (1092.00 g) gave 9.20, 14.34 and 37.55 g of dark viscous extracts after solvent removal, respectively (Table 1). While similar extraction on C. pubescens (1962.44 g) gave 15.25, 70.48 and 116.05 g of crude extracts, respectively.

Bioassay-guided isolation: These extracts were evaluated for antimicrobial activity by disc diffusion method. The hexane and chloroform extracts of C. impressicostatum and C. pubescens inhibited the growth of fungi strongly. Therefore, both the hexane and chloroform extracts were subjected to silica gel column chromatography separately. For the C. pubescens hexane extract, the column was eluted with gradient hexane and ethyl acetate and 83 fractions of 100 mL each were collected. Fractions showing similar TLC profiles were bulked to give F1 (fractions 1-7), F2 (7-15), F3 (15-19), F4 (19-26), F5 (26-34), F6 (34-41), F7 (42-45), F8 (46-51) F9 (51-55), F10 (55-66), F11 (67-73) and F12 (73-83). These fractions were tested for antimicrobial activity by disc diffusion method and the active fractions F2, F4, F5, F7, F8, F9 and F10 were determined (Table 2). Further chromatographic isolation of F4 and F5 yielded safrole (1, 28 mg) as yellowish viscous oil. The more active fractions (F8, F9 and F10) were further isolated by column chromatography. Both fractions F8 and F9 yielded (E)-methyl cinnamate (2, 60 mg) as white solid which on recrystallization gave white needles, m.p 36°C. Similarly, fraction F10 was further fractionated by silica gel chromatography eluted with hexane-ethyl acetate to afford 50 fractions. Fractions 22-32 and 34-40 were purified by preparative TLC (100% chloroform) to give another batch of (E)-methyl cinnamate (2, 72 mg) and (E)-piperonylprop-2-enal (3, 46 mg) as white solid.

Table 1:	The extracts of C. impressicostatum and C. pubescens

Table 2:	Inhibition zone (in mm) of various samples Cinnamomum pubescens against bacteria and fungi
Sa: Staphylococcus aureus; Se: S. epidermis; Bcr: Bacillus cereus; Pa: Pseudomonas aeruginosa; Bce: Burkholderia cepacia; Sc: Saccharomyces cerevisiae; Cl: Candida lipolytica; Ca: C. albicans; Mc: Microsporum canis

Fig. 1:	The component, (1) safrole, (2) (E)-methyl cinnamate, (3) (E)-piperonylprop-2-enal, (4) β-sitosterol, (5) (E)-piperonylprop-2-enol and (6) cinnamic acid

The later compound was recrystallized with chloroform to afford white shiny needle, m.p. 83-85°C (Parsons et al., 1993). β-Sitosterol (4, 55 mg) was obtained from fractions 42-48 as white needles (Fig. 1).

The chloroform extract of C. pubescens (10.0 g) was subjected to flash column chromatography using gradient mixture of hexane and ethyl acetate to give 130 fractions of 100 mL each. Based on TLC profiles, the fractions were bulked and selected fractions (F1-11, F12-20, F21-40, F41-60, F 61-78, F79-88 F89-130) were subjected to disc diffusion assay. The more active fractions (F12-20 and F21-40 were repeatedly separated by column chromatography and preparative TLC and eluting with various mixture of hexane and ethyl acetate. Safrole (1, 15 mg), (E)-methyl cinnamate (2, 47 mg) and (E)-piperonylprop-2-enal (3, 21 mg) were obtained from fraction F12-20. From fraction F21-40, β-sitosterol (4, 39 mg) and (E)-piperonylprop-2-enol (5, 16 mg) were obtained as brownish oily liquid. The various UV, IR, MS, NMR spectral data of the compounds were obtained, analyzed and compared with reported data (detailed spectral data could be obtained from the author).

Bioassay-guided isolation work on the hexane extract of C. impressicostatum was also carried out with the isolation of similar antimicrobial (E)-methyl cinnamate (2) and three other components safrole (1), (E)-piperonylprop-2-enal (3) and cinnamic acid (6) (Fig. 1).

RESULTS

The growth of the four fungi species were strongly suppressed by the hexane extracts of C. pubescens with inhibition zone range from 15.0 to 20.4 mm. However, the extract only exhibited moderate to weak activity towards bacteria. Similar trend was observed for the chloroform extract. Both the hexane and chloroform extracts of C. impressicostatum were strongly active when tested against three of the fungi, Candida lipolytica, C. albicans, Microsporum canis, but moderately active and inactive towards Saccharomyces cerevisiae, respectively. Moderate inhibition activity was also observed for chloroform extract against the bacteria. By using bioassay-guided isolation technique, the hexane and chloroform extracts of extracts C. pubescens furnished six compounds with (E)-methyl cinnamate as the most active component and the structure of the compounds were elucidated by detail spectroscopic method and comparison with literature values. For the pure isolated (E)-methyl cinnamate, the compound displayed excellent activity against the tested fungi with inhibition zone range from 24.5 to 19.8 mm. (E)-Piperonylprop-2-enal (3) only gave very weak inhibition on the growth of two of the bacteria and one of the fungi. β-Sitosterol (4) failed to exhibit any activity to both microbes (Table 2).

DISCUSSION

Both the hexane and chloroform extracts of barks of C. impressicostatum and C. pubescens were found to be active against all microbes tested especially towards fungi S. cerevisiae, C. albicans, C. lipolytica and M. canis by using disc diffusion method (Table 2). The activities are comparable with nystatin standard used. However, compared to standard streptomycin, the activities of the extracts against bacteria were much weaker. In the case of methanol extract, it is only selectively active and the quantum of activity is much lower. Having established the active extracts, the hexane and chloroform extracts of both plants were selected for further chromatographic isolation. Selected fractions from these were subjected to bioassay-guided isolation work. Further isolation of fractions F4 and F5 yielded safrole (1) as yellowish oil. Based on GC and GC-MS analysis, fractions F8, F9, F10, F12-20 and F21-40 were found to contain one major component. Bioassay-guided fractionation of fraction F8, F9 and F10 led to the isolation of four compounds (2, 3, 4), with compound (2) as the major component.

The spectral data of (1) revealed that the compound was safrole previously isolated from many plant species including Cinamomum petrophilum, C. mollissimumand Piper hispidinervum (Rocha and Lin, 1999). The compound was weakly active towards three of the fungi. The major component of the extract was compound (2) and the spectral data established the structure as (E)-methyl cinnamate (Bauer et al., 1990). The compound was isolated from both plant extracts and exhibited strong inhibition on the growth of all four fungi S. cerevisiae, C. albicans, C. lipolytica and M. cani with inhibition zone ranging from 24.5 to 19.8 mm. The compound also suppressed the growth of one of the bacteria (Bacillus cereus) with inhibition zone of 18.8 mm. These results are comparable with nystatin used as standard for antifungal property. The minimum inhibitory concentration of the compound using the broth microdilution method was also determined and it displayed MIC values of 1.25 μg μL^-1 for all the fungi tested and 5.00 μg μL^-1 for B. cereus. While we were preparing this manuscript, another article appeared which reported similar findings on the antimicrobial activity of commercial (E)-methyl cinnamate, but the test was carried out against E. coli, B. subtilis, S. aureus and C. albicans (Huang et al., 2009). In the current study, similar result was obtained towards C. albicans, but the compound was found to be weakly active against S. aureus. (E)-methyl cinnamate can be found it many other plant species and used in many application, as an ingredient in decorative cosmetics, fine fragrances, shampoos, toilet soaps and many other toiletries as well in non-cosmetic products (Rosas et al., 2005; Gupta, 1996; Bhatia et al., 2007). Various studies have been conducted to assess the toxicological and dermatological properties of the compound as an ingredient in fragrances (Belsito et al., 2007). The third compound isolated was identified as (E)-piperonylprop-2-enal (3) by analyzing the spectral data acquired and compared with literature values. It was obtained as white needles with melting point of 83-85°C (Parsons et al., 1993, m.p. 84-85°C). The compound was only weakly capable of suppressing the growth of certain bacteria and fungi, while β-sitosterol (4) failed to exhibit any activity at all. β-Sitosterol (4) is widely distributed in many plant species and can be obtained quite easily from their extracts.

Chromatographic separation of most active fractions F12-20 and F21-40 from the chloroform extract afforded the same four compounds described above and a brownish oily liquid established as (E)-piperonylprop-2-enol (5). Similarly, the structure of the compound was elucidated by spectral data analysis. This is a reduced form of the above aldehyde (3). Due to insufficient amount of sample isolated, the growth inhibition property of (3) against the microbes was not performed. Bioassay-guided chromatographic separation work was also conducted on the hexane extract of the leaves of C. pubescens to afford (E)-methyl cinnamate (2) as the most active component together with safrole (1), (E)-piperonylprop-2-enal (3) and cinnamic acid (6). The occurrence of (E)-piperonylprop-2-enal (3) was reported for the first time in the genus Cinnamomum although it has been isolated and identified previously from Brombya platynema. The assay results showed that (E)-methyl cinnamate (2) contributed to the overall antifungal activity of the hexane and chloroform extracts.

ACKNOWLEDGMENTS

We wished to thank Universiti Putra Malaysia for providing the facilities and the Government of Malaysia for providing financial support under the IRPA Program. One of us (N.A.M. Ali) wished to acknowledge Forest Research Institute of Malaysia for granting study leave to pursue a PhD degree.