Bioactivity-guided isolation and structural characterization of the antifungal compound, plumbagin, from Nepenthes gracilis

Abstract

Context: Despite several phytochemical studies of Nepenthes gracilis Korth (Nepenthaceae), the biological activities of this pitcher plant remain to be explored.

Objective: This study evaluates the antifungal activity of N. gracilis extracts, isolates, and characterizes its bioactive compound and evaluates the cytotoxicity of the isolated compound.

Materials and methods: Fresh leaves of N. gracilis were sequentially extracted. The fungistatic and fungicidal activities of the extracts were evaluated against six species of fungi of medical importance using a colorimetric broth microdilution method. The most active extract was fractionated by liquid–liquid partitioning and further purified by a preparative thin layer chromatography. Structural elucidation was carried out using FT-IR, GC-MS, and NMR. Cytotoxicity testing against rhesus monkey kidney epithelial cells (LLC-MK2) was assessed by a neutral red uptake (NRU) assay.

Results: The hexane extract, which showed the lowest minimum inhibitory concentration (MIC) and minimum fungicidal concentration (MFC), both at 20 μg/mL against Candida albicans, Issatchenkia orientalis, and Trichophyton mentagrophytes, was subjected to bioactivity-guided fractionation. The isolated compound exhibited potent activity with the MIC values ranging from 2 to 31 μg/mL against all the fungi. The active compound was identified as plumbagin (5-hydroxy-2-methyl-naphthalene-1,4-dione). The 50% cytotoxicity concentration (CC₅₀) of plumbagin was 0.60 μg/mL.

Discussion and conclusion: The selectivity indices of plumbagin against all the fungi were less than 1.0, indicating that plumbagin is more toxic to mammalian than fungal cells. This study provides information on the antifungal properties of N. gracilis leaf extracts, as well as the antifungal and cytotoxicity properties of plumbagin.

• Broth microdilution
• extraction
• fungicidal
• fungistatic
• Nepenthaceae
• naphthoquinone

Introduction

In the early 1900s, the pathogenic potential of certain endemic fungi was first recognized when most of the systemic fungal diseases were described. A dramatic rise in invasive mycoses resulting from alterations in immune status associated with the acquired immune deficiency syndrome (AIDS) pandemic, organ and bone marrow transplantations, and cancer chemotherapy (Kriengkauykiat et al., 2011; Pappas et al., 2010). These diseases cause morbidity and mortality in developed as well as developing countries as the availability of antifungal agents remains relatively limited and the emergence of widespread drug resistance (Denning & Hope, 2010; Miceli & Lee, 2011; Okeke et al., 2005). In addition, many antifungal drugs can have serious side effects due to the high degree of phylogenetic relatedness between fungi and humans as both are eukaryotes (Heitman et al., 2006). This poses great difficulties in developing drugs that target fungi without affecting human cells.

Nepenthes gracilis Korth. (Nepenthaceae) (Figure 1) is commonly known as “monkey cup”, which refers to the fact that monkeys occasionally drink rainwater from the pitchers of these plants. It grows in nutrient-poor soils with a shallow root system. It has highly modified epiascidiate sword-shaped leaves with tendrils that form the pitfall traps that efficiently capture, retain, and digest insect prey for nitrogen source (Elison & Gotelli, 2001). Pitcher plants are used in traditional medicine to treat dysentery, soothe inflamed skin, as an eye-wash, and as an astringent (Burkill, 1966). Extracts of Nepenthes mirabilis (Nepenthaceae) and Nepenthes ventricosa x maxima (Nepenthaceae) have been reported to possess antimicrobial activity against bacteria (Wiart et al., 2004) and plant pathogens (Shin et al., 2007a), respectively. No studies have been reported on the antifungal activity of N. gracilis. This study evaluated the antifungal property of N. gracilis leaf extracts. The active compound was isolated and characterized by high performance thin layer chromatography (HPTLC), Fourier transform infrared spectroscopy (FT-IR), gas chromatography-mass spectrometry (GC-MS), and nuclear magnetic resonance (NMR) spectroscopy.

Figure 1. Nepenthes gracilis Korth.

Materials and methods

Chemicals and reagents

The following chemicals and reagents used were of analytical grade purity: hexane (Mallinckrodt Chemicals, Phillipsburg, NJ), chloroform (System, Watsonville, CA), ethyl acetate (R&M, Hampshire, UK), ethanol and glacial acetic acid (PROCHEM, Alpharetta, GA), methanol (RCI Labscan, Bangkok, Thailand), amphotericin B, Dulbecco's phosphate buffered saline (DPBS), Dulbecco's modified Eagle medium (DMEM), neutral red solution (3.3 g/L in DPBS), fetal bovine serum (FBS), penicillin–streptomycin solution (100×), trypsin-EDTA solution (1×), sodium bicarbonate, sodium chloride and p-iodonitrotetrazolium violet (Sigma-Aldrich, St. Louis, MO), hexane and ethyl acetate of HPLC grade, deuterium oxide (D₂O), dimethyl sulphoxide (DMSO), glycerol, potassium bromide (KBr), potato dextrose agar (PDA) and sodium hydroxide pellets (Merck, Darmstadt, Germany), RPMI-1640 medium supplemented with glutamine and phenol red, without bicarbonate (MP Biomedicals, Strasbourg, France), and 3-(N-morpholino) propanesulphonic acid (MOPS) (Calbiochem, EMD Bioscience, La Jolla, CA).

Fungal strains tested

Candida albicans (ATCC 90028), Candida parapsilosis (ATCC 22019), Issatchenkia orientalis (ATCC 6258), Cryptococcus neoformans (ATCC 90112), Aspergillus brasiliensis (ATCC 16404), and Trichophyton mentagrophytes (ATCC 9533) were purchased from the American Type Culture Collection. The microorganisms were maintained on PDA at 4 °C.

Cell culture

The rhesus monkey kidney epithelial cell line (LLC-MK2) was a gift from Prof. Dr. Shamala Devi (Department of Medical Microbiology, Faculty of Medicine, University of Malaya, Malaysia). LLC-MK2 was propagated in DMEM supplemented with 5% FBS at 37 °C in a humidified atmosphere with 5% CO₂. For maintenance medium, the FBS concentration was reduced to 1%.

Plant materials

Nepenthes gracilis specimens were purchased on 26 September 2010 from a nursery in Cameron Highlands, Pahang, Malaysia. The identification of the species was confirmed by one of the co-authors, Hean Chooi Ong, an ethnobotanist from the University of Malaya. A voucher specimen of N. gracilis (UTAR/FSC/10/016) was deposited at the Faculty of Science, Universiti Tunku Abdul Rahman, Perak Campus.

Preparation of extracts

Fresh leaves were washed thoroughly with tap water to remove dirt and dust. The fresh leaves were weighed (237.23 g) and cut into small pieces, blended, and sequentially extracted using hexane, chloroform, ethyl acetate, ethanol, methanol, and distilled water at room temperature with agitation (120 rpm) using an orbital shaker (IKA-Werke KS 501, Staufen, Germany). Two cycles of extractions (1 d/cycle) were performed for each solvent. The solvents were filtered and evaporated using a rotary evaporator (Buchi Rota-vapor R205, Brinkman, Switzerland) at 40 °C. The water extracts were lyophilized using a freeze-dryer (Martin Christ Alpha 1-4, LD Plus, Osterode, Germany). The samples were then re-dissolved in a methanol–water mixture (2:1, v/v) at a concentration of 10 mg/mL for the crude extracts, 1 mg/mL for the isolated fractions, and 0.5 mg/mL for the isolated compounds. All dissolved samples were filtered using a 0.45 μm syringe filters and stored at −20 °C prior to analyses.

Preparation of fungal inocula

The preparation of broth medium and inoculum suspensions was based on NCCLS/CLSI guidelines (NCCLS, 2002a, 2002b). The inoculum suspensions were prepared from fresh, mature cultures grown on PDA. The suspensions were mixed using a vortex mixer (VELP® Scientifica, Milano, Italy) at 25 Hz for 15 s and adjusted to the optimal absorbance (A) values (A = 0.12−0.15 for Candida spp., I. orientalis and C. neoformans; A = 0.09 − 0.11 for A. brasiliensis, and A = 0.15 − 0.18 for T. mentagrophytes) at 530 nm. Further dilutions using sterile saline solution (0.9% NaCl) were performed to obtain the final working inoculum concentration (1–5 × 10³ CFU/mL for Candida spp. and I. orientalis; 1–5 × 10⁴ CFU/mL for C. neoformans; 0.4–5 × 10⁴ CFU/mL for A. brasiliensis, and 1.2–6 × 10⁴ CFU/mL for T. mentagrophytes).

Screening for antifungal activity

The colorimetric broth microdilution method of Eloff (1998) was modified and employed for screening the extracts for antifungal activity. This test utilized two-fold descending concentrations of the extracts in 96-well microplates and a reference antibiotic, with concentrations ranging from 2.5 to 0.02 mg/mL for the plant extracts, 0.25 to 0.002 mg/mL for the fractions, 125 to 0.98 μg/mL for the isolated compounds, and 8 to 0.063 μg/mL for amphotericin B. Growth, sterility, and negative controls for the extracts and medium were included. The 96-well microplates were incubated at 35 °C for 48 h for Candida spp. and I. orientalis; 72 h for C. neoformans and A. brasiliensis; and at 28 °C for 7 d for T. mentagrophytes (NCCLS, 2002a, 2002b). The colorimetric indicator, p-iodonitrotetrazolium violet (0.4 mg/mL), was added and a color change from colorless to red indicated a positive result. The concentration at which the color remains clear was recorded as the MIC value. The MFC was obtained by inoculating 20 μL of the preparation that showed no evidence of growth during the MIC determination assays on PDA. The lowest concentration at which growth was not observed was recorded as the MFC value. The tests were performed in triplicate.

Isolation of the active compound from N. gracilis

The most active extract, the hexane extract, was subjected to liquid–liquid partitioning. The dry hexane extract (1.12 g) was dissolved in 150 mL of hexane. This was partitioned with an equal volume of water. The water layer was then extracted using 150 mL of ethyl acetate. Each extraction was performed three-times and the extracts were pooled together. The hexane fraction and the ethyl acetate fraction were evaporated to dryness by rotary evaporation, while the water fraction was lyophilized. This yielded three fractions, i.e., hexane fraction (F₁), ethyl acetate fraction (F₂), and water fraction (F₃).

The dried samples for the F₁, F₂, and F₃ fractions were re-dissolved in a methanol–water mixture (2:1, v/v) for screening for antifungal activity. The F₁ fraction, which exhibited the strongest antifungal activity, was further purified by preparative thin layer chromatography (PLC) using a hexane–ethyl acetate mixture (9:1, v/v) as the mobile phase. The F₁ fraction was re-dissolved in hexane (2 mg/mL) and applied (190 µL) as a thin even layer horizontally above the solvent level on a 10 × 10 cm PLC plate. The development of PLC was performed at room temperature. Two bands (assigned as B₁ and B₂) were removed from the packing material (silica gel 60 F₂₅₄) of the PLC plate. The silica gels from those bands were extracted twice with distilled water, followed by partitioning with equal volume of hexane. The hexane solvent was collected and evaporated using a rotary evaporator and screened for antifungal activity. Following recovery of components in B₁, further purification was performed by re-crystallization in hexane. This yielded yellow crystals, which was identified as 5-hydroxy-2-methyl-naphthalene-1,4-dione (plumbagin) based on spectroscopic data as well as by comparison with the literature data.

Characterization of the active compound from N. gracilis

Melting point was measured in a glass capillary tube by using a melting point apparatus (Stuart SMP 10, Staffordshire, UK). FT-IR spectroscopy was performed on KBr pellets containing the sample (Perkin-Elmer Spectrum RX-1, Waltham, MA). HPTLC analysis and PLC separation were performed on silica gel plates (Kieselgel 60 F₂₅₄, 0.2 mm; aluminum-backed and 1 mm; glass-backed, respectively, Merck, Germany) using a semi-automatic sample applicator (Camag Linomat IV, Switzerland). The HPTLC plate was scanned using a TLC scanner (Camag TLC Scanner 3, Switzerland) and the data of absorption signal spectra (200–700 nm) were processed with WinCATS software (Planar Chromatography Manager version 1.4.4, Camag, Switzerland) for purity assessment. The HPTLC plates were photographed using a 12 bit charge-couple device (CCD) digital camera (Camag TLC Visualizer, Switzerland) under uniform illumination with white and UV light.

The mass spectrum of the isolated compound was obtained by GC-MS (Shimadzu, QP2010 Plus, Tokyo, Japan) using helium as the carrier gas. The working pressure was 99.8 kPa and the flow rate was 1.46 mL/min. The GC-MS was fitted with a 5% phenyl polysilphenylene-siloxane (BPX5) capillary column of dimension 0.25 mm × 30 m and the film thickness was 0.25 μm. The injector port was maintained at 250 °C while the oven temperature gradient was 10 °C/min from 80 °C to 310 °C with a total program time of 37 min. Electron impact ionization method was used. The ion source and interface temperatures were set at 200 °C and 310 °C, respectively. The constituent was identified by comparing with the mass spectral libraries (Wiley MS/NIST 05).

¹H-NMR (400 MHz) and ¹³C-NMR (100 MHz) spectra of the isolated compound were recorded in deuterium oxide (D₂O) on a NMR instrument (JMN-ECX 400, JEOL, Tokyo, Japan). All chemical shifts (δ) were stated in parts per million (ppm) with reference to tetramethylsilane (TMS) as the internal standard and coupling constants (J) were measured in Hz.

Determination of cytotoxic activity

LLC-MK2 cells (3 × 10⁴ cells/well) were seeded in 96-well microplates and incubated for 24 h at 37 °C in a humidified atmosphere with 5% CO₂. After incubation, the cells were treated with two-fold serially diluted concentrations of plumbagin (0.05–6.25 μg/mL) for 72 h. The final concentration of the diluent (methanol–water mixture, 2:1, v/v) for plumbagin was applied at the non-toxic concentration (<0.25%, v/v). Cell growth, sterility, and negative controls were also included. The cell viability was evaluated using the NRU assay according to the protocol of Repetto et al. (2008) with modifications. The absorbance of the extracted neutral red dye was measured at 540 nm using a microplate reader (Tecan Infinite-M200, Männedorf, Switzerland). The experiment was performed in triplicate. The 50% cytotoxicity concentration (CC₅₀) of plumbagin on the cells was determined from the plot of cell viability against concentrations of plumbagin. The data were analyzed with one-way ANOVA using Statistical Package for the Social Sciences (SPSS) software (Version 15.0 for Windows, SPSS Inc., Chicago, IL) and the significance level was set at p < 0.05. The toxicity of plumbagin to the LLC-MK2 cells and the antifungal activity of plumbagin against each species of fungus were compared by using the selectivity index (SI) where SI = CC₅₀/MIC.

Results

Following initial extraction, the highest percentage of yield (w/w) was obtained from the ethanol extract (1.21%), followed by chloroform, hexane, ethyl acetate, methanol, and water extracts which gave 0.55, 0.47, 0.39, 0.21, and 0.05%, respectively. A total of 1.12 g of hexane extract was obtained from 237.23 g of fresh leaves (0.47% yield). Further fractionation of the hexane extract reduced to 0.783 g (0.33% yield) of hexane fraction (F₁), 0.247 g (0.10% yield) of ethyl acetate fraction (F₂), and 0.028 g (0.01% yield) of water fraction (F₃). Plumbagin (290 mg) was isolated from the hexane fraction (0.12% yield).

The antifungal activities of extracts and plumbagin isolated from N. gracilis against six types of fungi were evaluated using a colorimetric broth microdilution method. The minimum inhibitory concentrations (MIC) and minimum fungicidal concentrations (MFC) of various extracts and isolated plumbagin are shown in Table 1. The MIC values ranged from 20 to 2500 μg/mL for the six extracts, from 2 to 31 μg/mL for plumbagin, and from 0.125 to 8 μg/mL for amphotericin B (positive control). The ethanol, methanol, and water extracts did not exhibit any antifungal activity against the molds used (A. brasiliensis and T. mentagrophytes), while the hexane extract showed the lowest MIC value (20 μg/mL) against C. albicans, C. parapsilosis, I. orientalis, C. neoformans, and T. mentagrophytes. The hexane extract was, therefore, selected for fractionation and isolation of active compounds.

Table 1. MIC and MFC values of extracts from different solvents and isolated plumbagin from the leaves of N. gracilis against six types of fungi.

	MIC (μg/mL)								MFC (μg/mL)
Fungi	Hex	CH	EA	ET	MT	AQ	B1	Ref	Hex	CH	EA	ET	MT	AQ	B1
C. albicans	20	313–625	625	313	313	2500	2	2	20	625	NA	NA	NA	NA	2
C. parapsilosis	20	1250	625	313–625	2500	NA	8	4	160	2500	NA	NA	NA	–	16
I. orientalis	20	313	313	160	80–160	40–80	2	4	20	313	NA	NA	NA	NA	2
C. neoformans	20	313–625	313	160	40–80	40	4	0.125	160	1250	2500	2500	NA	NA	8
A. brasiliensis	40–80	1250	NA	NA	NA	NA	31	2	160	NA	–	–	–	–	63
T. mentagrophytes	20	313	1250–2500	NA	NA	NA	2	8	20	313	2500	–	–	–	4

The results are shown as the mean or range values of triplicates. Hex, hexane extract; CH, chloroform extract; EA, ethyl acetate extract; ET, ethanol extract; MT, methanol extract; AQ, water extract; B1, plumbagin; Ref, reference drug, amphotericin B; NA, no activity; “–”, not tested since no MIC value was obtained.

Cryptococcus neoformans and T. mentagrophytes were chosen as indicative microorganisms for the bioactivity-guided fractionation. Following fractionation of hexane extract, F₁ fraction exhibited the lowest MIC and MFC values against the indicative microorganisms, ranging from 8 to 16 μg/mL, followed by F₂ fraction at 16–31 μg/mL while F₃ did not show any antifungal activity (Table 2). The F₁ fraction contained the active compounds. Further isolation of this fraction using preparative thin layer chromatography (PLC) yielded two compounds (B₁ and B₂). The concentration of B₁ compound that inhibited the growth of the indicative microorganisms was about 32-fold lower than that required for compound B₂ (Table 2). Compound B₁ was thus subjected to further characterization. Compound B₁ effectively inhibited the growth of all tested fungi and exerted a fungicidal effect on all of them, with the MFC values ranging from 2 to 63 μg/mL (Table 1). C. albicans and I. orientalis were the most susceptible species with the lowest MFC value of 2 μg/mL.

Table 2. MIC and MFC values of different fractions and compounds isolated from F₁ of the hexane extract of N. gracilis.

	MIC (μg/mL)		MFC (μg/mL)
Fraction/compound	T. mentagrophytes	C. neoformans	T. mentagrophytes	C. neoformans
Hexane fraction (F1)	8	16	8	16
B1	2	4	4	8
B2	62.5	125	125	NA
Ethyl acetate fraction (F2)	16	31	16	31
Water fraction (F3)	NA	NA	–	–
Ref	8	0.125

The results are shown as the mean values of triplicates. The meaning of the abbreviations are as follows: B1, 1st compound isolated from F₁; B2, 2nd compound isolated from F₁; Ref, reference drug, amphotericin B; NA, no activity; “-”, not tested since no MIC value was obtained.

B₁ compound was subjected to HPTLC, FT-IR, GC-MS, and NMR analyses, and was subsequently identified as plumbagin (5-hydroxy-2-methyl-naphthalene-1,4-dione). Plumbagin forms yellow needle-shaped crystals and has a melting point of 78–79 °C. It had an R_f of 0.34 on a HPTLC silica gel 60 F₂₄₅ plate developed with hexane–ethyl acetate (9:1, v/v) as the mobile phase. The UV–Vis absorption spectrum displayed two absorption maxima at 260 nm and 425 nm, identical to those reported by Bothiraja et al. (2011). The IR spectrum (KBr) showed the presence of –OH stretching at 3450 cm⁻¹, conjugated carbonyl groups at 1640 cm⁻¹, and the aromatic C–H stretching at 803 cm⁻¹. The GC-MS spectrum showed a molecular ion peak at 188.05 m/z and the spectrum obtained matched with the reference standard plumbagin reported by Shin et al. (2007b). The ¹H- and ¹³C-NMR spectra confirmed the presence of plumbagin with the molecular formula C₁₁H₈O₃.

The viability of LLC-MK2 cells treated with various concentrations of plumbagin as measured by NRU assay is given in Figure 2. Plumbagin was found to be significantly cytotoxic (p < 0.05) to the cells at concentrations ranging from 0.78 to 6.25 μg/mL with the cell viability values less than 10% after 72 h treatment. According to Figure 2, the CC₅₀ value of plumbagin on the LLC-MK2 cells was 0.60 µg/mL (3.19 μM). The cytotoxicity was not observed at concentrations below 0.20 μg/mL as the percentages of cell viability were approximately 100%. The SI values calculated were 0.30 for C. albicans, I. orientalis, and T. mentagrophytes, 0.15 for C. neoformans, 0.08 for C. parapsilosis and 0.02 for A. brasiliensis.

Figure 2. Cytotoxic effects of isolated plumbagin from N. gracilis on rhesus monkey kidney epithelial cells (LLC-MK2). LLC-MK2 cells were treated with various concentrations (0.05–6.25 μg/mL) of plumbagin for 72 h. The results represent mean ± standard deviation of three replicates. *Statistically significant (p < 0.05) with one-way ANOVA.

Discussion

Several phenolic constituents, such as plumbagin, isoshinanolone, epishinanolone, shinanolone, quercetin, kaempferol, and flavonoid gallate esters, have been isolated from the leaves of N. gracilis (Aung et al., 2002; Fan et al., 2010). However, reports of the pharmacological and biological investigation of N. gracilis are scarce. To the best of our knowledge, the present study is the first report to investigate the antifungal property of N. gracilis leaf extracts against fungi of medical important.

The six extracts obtained from the leaves of N. gracilis exhibited antifungal activity against at least three types of fungi (Table 1). The bioactivity-guided approach led to the isolation of plumbagin from the hexane extract of N. gracilis. Plumbagin is a secondary metabolite belonging to the naphthoquinone group which is produced by many plants belonging to the Caryophyllales order (Richer et al., 2002) such as Droseraceae (Budzianowski, 2000), Plumbaginaceae (Bothiraja et al., 2011), Ebenaceae (Evans et al., 1999), and Nepenthaceae (Aung et al., 2002). A wide range of biological properties of plumbagin such as anticancer (Kawiak et al., 2007), antiviral (Perez-Sacau et al., 2003), anti-inflammatory, antiplatelet, antiallergic (Lien et al., 1996), antimalarial (Biot et al., 2004), and antibacterial (Cai et al., 2000) have been reported. Dzoyem et al. (2007) reported the antifungal activity of plumbagin isolated from the stem bark of Diospyros crassiflora. In this study, the concentration of plumbagin that inhibited the growth of I. orientalis, and T. mentagrophytes was about 2-fold and 4-fold lower, respectively, than that required for amphotericin B (Table 1).

The values of the selectivity indices of the isolated plumbagin against all the fungi were less than 1.0, indicating plumbagin is more toxic to mammalian than fungal cells. Seshadri et al. (2011) have shown that the cytotoxic effect of plumbagin in human peripheral blood lymphocytes was at least two-fold higher than that of juglone (5-hydroxy-1,4-naphthalenedione). Plumbagin at 5 µM was proven to be non-cytotoxic to the resting mouse lymphocytes, but at the concentration of 2 µM was found to induce apoptosis in human resting lymphocytes, indicating the species-specific differences in the activity of plumbagin (Checker et al., 2009). Two major mechanisms have been proposed for plumbagin cytotoxic action in various biological systems. First mechanism is associated with the excessive generation of reactive oxygen species (ROS) such as superoxide radicals, singlet oxygen, and hydrogen peroxide (Castro et al., 2008; Seung et al., 1998). Second mechanism involved the redox and oxidation cycle of quinones namely “redox cycles” which lead to their ability to act as potent electrophiles to inhibit the electron transportation, as uncouplers of oxidative phosphorylation, as intercalating agent in the DNA double helix and as bio-reductive alkylating agents in the biomolecules (Babula et al., 2009). The redox cycling and free radical forming was identified as the main mode of actions for plumbagin.

In animal studies, it has been noted that treatment with plumbagin causes reproductive toxicity (Bhargava, 1984; Premakumari et al., 1977). The oral bioavailability of plumbagin in rat was less than 40%, due to its high lipophilicity and insolubility in water (Hsieh et al., 2006). Consequently, a large and frequent dose is necessary to achieve the optimum therapeutic efficacy and this may lead to severe side effects. The toxicity of plumbagin is a major obstacle to its clinical application. The data reported herein are important, taking into account the medical importance of the studied fungi and that this plant species is traditionally consumed or applied externally, without considering the presence of toxic components such as plumbagin. Thus, the pharmacological study of medicinal plants remains important to provide evidence with scientific basis for the continued traditional application of plants and understanding of a plant's efficacy as well as toxicity.

Conclusions

Plumbagin plays a significant role in the antifungal activity of N. gracilis. The present study demonstrated that the leaves of this plant could serve as an antifungal agent but caution should be taken as plumbagin is a potentially cytotoxic compound.

Declaration of interest

The authors declare no conflict of interest.

Acknowledgements

The authors thank Universiti Tunku Abdul Rahman for a research grant (UTARRF Vote No. 6200/S07) which supported the M.Sc. candidature of Pei Shing Gwee.