MATERIALS AND METHODS

Instrumentation: This research work was conducted between 2009 and 2012. The IR spectra were recorded on a Perkin Elmer FTIR 1650 spectrophotometer using KBr discs measured in cm^-1; MS were recorded on a SHIMADZU GC-MS QP2010 Plus with EI electron impact ion source of 70eV; ¹H-NMR ¹³C-NMR, DEPT-135, HMQC, HMBC, COSY NMR spectra were obtained on a Varian AS 400 spectrometer using CDCl₃ as solvent; High Performance Flash Chromatography (HPFC) was performed using a Biotage Inc. Horizon pump (Charlottesville, VA) equipped with a Horizon flash collector and a fixed wavelength (254 nm) detector (system from Biotage Inc. Dyax Corp.). Biotage columns (surface area 500 m² g^-1, porosity 60 Å, particle size 40-63 μm) used were made up of FLASH 12+M, KP-SiL (12x150 mm) and FLASH 25+M, KP-SiL (25x150 mm); TLC was carried out on commercial TLC plates with fluorescent indicator (250 μm, silica gel GF Uniplate, Analtech Inc., Newark DE), developing with solvent mixture made up of EtOAc and hexane, visualization either under UV light at 254 or 365 nm and/or by spraying with anisaldehyde and heating at 110°C.

Statistical analysis: The treatments were arranged in a completely randomized design in four replicates with Allamanda spp., types of extracts and concentration levels as factors. All data were subjected to analysis of variance (ANOVA) where significant (p<0.05) differences between means were determined by Tukey’s standardized range test (HSD). Relationships (quadratic) between colony diameter and extract concentration were determined by regression analysis of data obtained for the different Allamanda species and types of extracts. The SAS (version 9.2) software was used to perform all analyses.

Plant materials: Allamanda blanchetti, Allamanda cathartica, Allamanda cathartica ‘Alba’, Allamanda cathartica ‘Jamaican Sunset’ and Allamanda oenotheraefolia were obtained from nurseries in Sungai Buloh, Selangor and University Agriculture Park, UPM Serdang, Malaysia. The leaves were washed, air-dried at room temperature, ground into powder and stored until required for extraction.

Plant extraction and bioassay: The dried fine powdered leaf of each plant (1 kg) was separately and sequentially extracted with petroleum ether, chloroform and methanol at room temperature. The extracts were filtered through Whatman No. 1 filter paper and concentrated by rotary evaporator to give fifteen dark sticky semisolid extracts. The stock solutions (100 mg mL^-1) were serially diluted with acetone to obtain the desired concentration for test solutions of 1, 3, 5 and 7 mg mL^-1. Effect of Allamanda extracts on mycelia growth of C. gloeosporioides was measured in vitro according to the method described by Bautista-Banos et al. (2003) using poison agar technique. PDA was amended with 4 concentrations of each extract (1, 3, 5 or 7 mg mL^-1). Fifteen mL of the amended PDA was poured into petri dishes (9 cm) and a fungal plug (4 mm) from pure culture of C. gloeosporioides was placed in the centre of the dishes. Petri dishes were then incubated at room temperature and radial measurements of growth were recorded daily until the fungus reached the edge of the plate. Petri dishes containing PDA and acetone served as control. Percentage Inhibition of Radial Growth (PIRG) was calculated after 8 days of incubation using a standard formula:

where, R₁ is the radial growth of fungus in the control plate (mm), while R₂ is the radial growth of the fungus containing extracts (mm) (Sivakumar et al., 2000).

Bioautography-guided fractionation and isolation of plumericin: The chloroform extract of A. cathartica ‘Jamaican Sunset’ (5 g) was absorbed into silica gel Biotage samplet and applied to a Biotage column FLASH 25+M, KP-SiL (150x25 mm i.d.; 40 mL min^-1). Elution of the column was performed by using mixtures of EtOAc:hexane (1:9, 3:7; 1:1; 7:3; 9:1) to give 240 fractions of 50 mL each. Based on TLC analysis, similar fractions were combined and their activity was monitored by bioautography. The active combined fractions 86-112 were further separated by preparative TLC (20x20 cm) and eluted twice with ether:hexane (1:1) and the two active bands were scraped off separately. The scraped bands were dissolved in dichloromethane, filtered through cotton wool to give two white solids. Due to insufficient amount of one of the samples, only the major and active component was submitted for further spectroscopic analysis and identified as plumericin. IR v_max cm^-1 (CHCl₃): 2921, 2854 (CH), 1759 (C = O), 1696 (C = O), 1448 (C = C); ¹H-NMR (CDCl₃, 400 MHZ)-see Table 1; ¹³C-NMR (CDCl₃, 400 MHZ)-see Table 1; EIMS m/z (%): 290 (M⁺) (35), 272 (30), 258 (60), 230 (100), 212 (20), 193 (60), 173 (45), 160 (70), 145 (60), 139 (85), 115 (55), 105 (40).

Table 1:	1H-NMR (400 MHZ, CDCl3) and 13C-NMR (100 MHZ, CDCl3) spectral data of plumericin
*Brigitta et al. (2005)

Bioautography: TLC plates for bioautography assays were spotted with 80 μg of crude plant extracts and selected fractions and then chromatographed in duplicate (Fig. 1). Fungicide technical grade standards benomyl, cyprodinil, captan (Chem Service, Inc., West Chester, PA) and azoxystrobin (Syngentia, Greensboro, NC) were used as controls at 2 μg in 4 μL of acetone. TLC plates were sprayed with spore suspension adjusted to a final concentration of 1.0x0⁶ conidia mL^-1 with liquid potato dextrose broth (PDB, Difco) and 0.1% Tween-80 in order to detect biological activity directly on the TLC plate. Using a 50 mL chromatographic sprayer, each TLC plate with a fluorescent indicator was sprayed lightly with the conidial suspension. Inoculated plates were then placed in a 30x13x7.5 cm moisture chamber (25°C, 100% relative humidity, Pioneer Plastics, Inc., Dixon, KY) and incubated in a growth chamber at 24±1°C. Inhibition of fungal growth was measured 4 days after treatment. The sensitivity of the fungus to each test sample was indicated by the presence of clear zones of fungal growth inhibition directly on TLC plates.

RESULTS AND DISCUSSION

The petroleum ether, chloroform and methanol extracts of the five plants were monitored for antifungal activity by poison agar technique and the three chloroform extracts of A. blanchetti (CEB), A. cathartica ‘Alba’ (CEA) and A. cathartica ‘Jamaican Sunset’ (CEJS) with strong antifungal activity were selected for further investigation. TLC development of these extracts eluted with ethyl acetate:hexane (3:7) revealed the presence of several prominent spots in each plate with the same R_f values range from 0.17-0.77 (Fig. 1a). The plates were then sprayed with C. gloeosporioides and incubated for 4 days at 24±1°C.

Fig. 1(a-b):	(a) TLC chromatograms and (b) Bioautograms of A. cathartica ‘Alba’ (CEA), A. blanchetti (CEB) and A. cathartica ‘Jamaican Sunset’ (CEJS) chloroform extracts against C. gloeosporioides in 30% ethyl acetate:hexane as solvent system. (Clear zones indicate inhibition of fungal growth by compounds in the extracts after 4 days of incubation at 24±1°C

Very clear and prominent zones were visible on three of the active spots with the same R_f values (0.3-0.55) on the bioautograms (Fig. 1b). This clearly indicated the presence of phytochemicals which inhibited the growth of C. gloeosporioides. The chloroform extract of A. cathartica ‘Jamaican Sunset’ (CEJS) was selected for larger scale separation by using High Performance Flash Chromatography (HPFC) with Biotage column FLASH 12+M, KP-SiL (12x150 mm) and eluted with various mixtures of solvents to give 240 fractions of 50 mL each. All fractions collected were monitored by bioautography and the active fractions (86-112) with prominent clear zones were combined and further purified by preparative TLC.

Fig. 2:	Structure of plumericin

The major and prominent active spot was scraped, dissolved in dichloromethane, filtered and on solvent removal white solid was obtained. Since the R_f values of the active spots for the two other chloroform extracts of A. blanchetti and A. cathartica ‘Alba’ were identical to the isolated compound, no further separation work was carried on them.

Confirmation of the structure of the isolated compound (Fig. 2) was carried out by comparing the spectral data obtained with literature values. The IR spectrum revealed existence of two carbonyl groups as indicated by the occurrence of prominent absorptions at 1696 and 1759 cm^-1. The mass spectrum showed a molecular ion peak at m/z 290 which corresponded to the molecular formula C₁₅H₁₄O₆ and the base peak occurred at m/z 230 due the cleavage of -CO₂CH₃. The integration of ¹H-NMR indicated the presence of fourteen protons which further supported the suggested molecular formula (Table 1). The two methyl groups observed at δ 2.08 (day) and 3.77 (sec) are assigned to C-14 and C-16, respectively. The rest of the proton signals are due to the occurrence of eight methine groups and they are assigned based on coupling constants, COSY and HMBC correlations spectra. Two of them are observed as singlets at δ 7.44 (H-3) and 5.10 (H-10). Another two occurred as doublets at δ 5.56 and 6.04 assigned for H-1 and H-7. A quartet observed at δ 7.16 which exhibited ²J correlation with the carbonyl group (C-12) is assigned to H-13. The ²J and ³J correlations between protons and carbons observed in the HMBC spectrum further supported the assignments of other methine groups. The attachment of the carboxylate group to the pyran ring is rationalized by the correlation of H-3 to the carboxyl carbon. The existence of fifteen carbon atoms displayed by the ¹³C-NMR and DEPT spectra are made up of two methyls, eight methines, three quarternary carbons and two carbonyl groups. The assignments of various protons to their respective carbon atoms could be clearly seen in the HMQC spectrum. Based on these detailed spectral analysis and comparison with literature report, the compound is identified plumericin previously isolated and identified from Plumeria rubra (Brigitta et al., 2005).

Plumericin is an important member of tetracyclic sesequiterpene commonly found in various plant families including from different genera of Apocynaceae such as Plumeria, Allamanda and Himatanthus (Silva et al., 2010; Sharma et al., 2011). The antifungal, anticancerous, antiviral and antibacterial actions of plumericin have been reported previously (Singh et al., 2011; Sticher, 1977). The compound isolated from another species within the Apocynaceae (Himatanthus sucuuba) has been demonstrated to have antiparasitic activity against Leishmanial amazonnensis, responsible for cutaneous leishmaniasis (Castillo et al., 2007). It is believed that the strong activity of the tetracyclic plumericin is due the presence of an á-methylene-γ-lactone moiety which is susceptible to undergo a Michael-type addition with biological nucleophiles (Kupchan et al., 1970). Knobloch et al. (1986) reported that sesquiterpenes affect the bacterial processes that include the inhibition of electron transport, protein translocation, phosphorylation steps and other enzyme-dependent reactions. Most probably similar mechanisms of action were responsible for the antifungal activity of the compound under study. Structural characteristics and chemical properties of terpenes and fatty acids played major roles in signifying antifungal activity for compositions containing these compounds with the presence of oxygenated substituent such hydroxyl and carboxyl groups (Glinski and Branly, 2002).

CONCLUSION

Antifungal-guided isolation of the chloroform extract of A. cathartica ‘Jamaican Sunset’ has led to the identification of plumericin as the compound responsible for the potent antifungal activity exhibited by the plant. The structure of the compound was determined by detailed spectroscopic analysis and comparison with previous reports and has the potential to control anthracnose disease caused by pathogenic fungus Colletotrichum gloeosporioides, an economically important disease in postharvest papaya cultivation. This study adds another important development of this interesting compound which has been the target of various chemical syntheses.

ACKNOWLEDGMENT

We gratefully acknowledge Dr. S.O. Duke, Dr. K.M. Meepagala, Mr. J. Martin and Ms. L. Robertson from USDA-ARS-NPURU for technical assistance. We also would like to acknowledge Universiti Putra Malaysia for providing facilities to carry out this project.