Cytotoxic and antioxidant potentials of ellagic acid derivatives from Conocarpus lancifolius (Combretaceae)

Purpose: Isolation, characterisation and structure elucidation of compounds obtained from Conocarpus lancifolius and screening of their pharmacological effects in vitro. Methods: After collection, authentication and extraction from whole C. lancifolius plants, screening for secondary metabolites, thin-layer chromatography and subsequent open column chromatography were performed for phytochemical analysis and subsequent purification of the compounds. The chemical structures of the isolated compounds were elucidated using spectroscopic (UV-visible, infrared and mass) spectroscopy, and nuclear magnetic resonance (1H-NMR, 13C-NMR including BB, DEPT-135, 90 and two-dimensional correlation techniques, including HMBC and HSQC). The cytotoxic and antioxidant potentials of extracts and compounds obtained from C. lancifolius were evaluated using in vitro models. Results: Two ellagic acid derivatives, 2,3,8-tri-o-methylellagic acid (A) and 3-O-methylellagic acid 4-Oβ-D-glucopyranoside (B), were isolated. Both compounds (A and B) were cytotoxic in a variety of cancer cell lines, including murine lymphocytic leukaemia (P-388, half-maximal inhibitory concentration (IC50) = 3.60 and 2.40 μg/mL, respectively), human colon cancer (Col-2, IC50 = 0.76 and 0.92 μg/mL, respectively) and human breast cancer (MCF-7, IC50 = 0.65 and 0.54 μg/mL, respectively). Moreover, both compounds showed significant antioxidant potential in vitro. Conclusion: C. lancifolius extract and isolated ellagic acid derivatives (compounds A and B) possess cytotoxic and antioxidant properties. These findings suggest that C. lancifolius contains bioactive compounds that can be potentially developed as natural cytotoxic and antioxidant compounds.


INTRODUCTION
Combretaceae is a medicinally important flowering plant family that comprises 20 genera.
Of these genera, Terminalia and Combretum have been extensively studied for their chemical composition and biological attributes. Combretaceae includes ornamental trees with a high-tolerance to heat and semi-arid conditions; they are usually grown in Saudi Arabia [1]. The leaves are simple and complete, spirally arranged or alternate having glands and domatia. The flowers are bisexual or male, 4-5-merous, actinomorphic, in axillary spikes or racemes. The genus Conocarpus consists of only two species, namely Conocarpus erectus and Conocarpus lancifolius [2]. C. lancifolius, an ornamental tree, is native to coastal and riverine areas of East Africa [3]. This fast-growing tree thrives in sandy soils and semi-arid conditions. Mature leaves are glossy in appearance with relatively few trichomes on both surfaces and contain two cavities or secretory ducts that secrete epicuticular waxes, polyphenols and polysaccharides [4]. A methanol extract of the aerial part of C. lancifolius has antidiabetic potential via suppression of gluconeogenesis in an alloxan-induced diabetes rabbit model [5]. The methanol extract of C. lancifolius fruit demonstrates a noticeable cytotoxic effect against MRC-5 cells, as well as antiprotozoal activity [6]. The alkaloidal extract of C. lancifolius leaves exhibits antibacterial activity [7]. The methanol extract of C. lancifolius shows moderate antibacterial and low antifungal activities [7]. Antioxidant, phytotoxic and antiurease activities, and total phenolic and flavonoid contents of C. lancifolius have already been reported by our research group [8].

EXPERIMENTAL
Plant material was collected from different areas Lahore (Pakistan) in August 2012. After authentication by Professor Dr Altaf Ahmad Dasti, Institute of Pure and Applied Biology, Bahauddin Zakariya University, Multan, Pakistan, a voucher specimen (no. WCL-291) was allotted and the sample was deposited in the institute's herbarium.

Materials
Materials used for extraction, chemical characterisation and isolation purposes were polar and non-polar solvents, including chloroform (C), dichloromethane (DCM), methyl alcohol (ME), ethyl acetate (EA), n-hexane (n-Hex) and vanillin. M TLC plates (Merck Silica gel 60 F254; 20 x 20 cm) were used for the separation of fractions.

Extraction
Whole C. lancifolius plants were dried and ground to form a coarse powder using a grinder. The coarse powder was subjected to sequential DCM and ME extraction by the successive maceration method. Soaked material was filtered through filter paper and the collected filtrates were concentrated using a rotary evaporator. The resultant extracts were labelled as CLD and CLM and were subjected to cytotoxic screening. Based on the half maximal inhibitory concentration (IC50), the CLM was selected for further fractionation and isolation of bioactive compounds.

Screening for secondary metabolites
The powdered plant material was weighed, and different qualitative tests for alkaloids, cardiac glycosides, anthraquinones, saponins, tannins and triterpenoids were performed following standard protocols.

Purification of compounds A and B
The CLM (10 g) was fractionated by step-wise elution using column chromatographic techniques. The stationary phase was silica gel 60 (40-63 μm) and the mobile phases was a mixture of C, ME and water. This protocol resulted in the collection of four fractions (CLM1-4). CLM-2 (5.2 g) was further fractionated through column chromatography using the mobile phase with an 80:20:2 ratio of C:ME:water and silica gel 60 with a pore size of 40-63 μm as the stationary phase. This procedure resulted in the collection of six sub-fractions (CLM-2a to CLM-2f). CLM-2b was collected as a pure compound A (42 mg). The CLM-2c (720 mg) sub-fraction was further fractionated. Seven subfractions were collected, out of which CLM-2c7 was a pure compound B (28 mg).

Evaluation of cytotoxic potential
The cytotoxic effect of both extracts (CLD and CLM) and isolated compounds (A and B) were studied using the in vitro sulforhodamine B (SRB) method [9]. Ellipticine and medium containing the same per cent of dimethyl sulfoxide (DMSO) were used as the positive and negative control, respectively. Extracts and isolated compounds were dissolved in DMSO to prepare stock solutions (4 mg/ml). Six different concentrations of test samples were prepared and tested in triplicate; the final DMSO concentration was 0.5%. Cell lines (P-388, MCF-7, ASK, HEK293 and T24) were cultured in Minimum Essential Medium (MEM) containing L-glutamine and 10% foetal bovine serum (FBS). Lu-1 cells were grown in MEM with 5% FBS.
After 48-(P-388) or 72-h treatment, cells were fixed with trichloroacetic acid (10%) and stained with sulforhodamine B (0.4% in 1% acetic acid). Unbound dye was removed by washing, while bound and dried stain was solubilised with 10 mM Trizma base. The absorbance was measured at 510 nm using a BMG FLUostar OPTIMA plate reader. The data are expressed as the median effective dose (ED50) ± standard error of the mean (SEM) of three independent experiments (n = 3).

Evaluation of antioxidant activity
Two in vitro methods, viz, 2,2-diphenyl-1picrylhydrazyl (DPPH) and nitric oxide (NO) free radical scavenging assays were employed to screen the antioxidant potential of the test samples.

DPPH free radical scavenging assay
In brief, 100 µL of DPPH solution (200 µM in methanol) was mixed with 100 µL of test solution (10 mg/ml CLM or 10 µM compound A or B). The reaction was allowed to proceed for 30 min at room temperature in the dark. The absorbance of the remaining DPPH free radical was measured using a multi-plate reader (Spectra Max 340) at 517 nm. The per cent radical scavenging activity (%RSA) was determined by comparison with a DMSO-containing control. The IC50 values of the compounds were calculated using EZ-Fit Enzyme kinetics software (Perrella Scientific Inc. Amherst, MA, USA). Ascorbic acid was used as a reference standard [10].

NO scavenging assay
In brief, 10 mM of sodium nitroprusside in phosphate-buffered saline was mixed with different concentrations of CLM or compounds A and B and incubated at 300°C for 2 h. The same reaction mixture without extract, but containing an equivalent amount of ethanol, was used as the control. After the incubation period, 500 µL of Griess reagent (1% sulfanilamide, 2% H3P04 and 0.1% N-(1-naphthyl) ethylenediamine dihydrochloride) was added to the reaction mixture. The absorbance of the chromophore formed during diazotisation of nitrite with sulfanilamide and subsequent coupling with naphthyl ethylenediamine dihydrochloride was measured instantly at 550 nm using a plate reader. The inhibition of nitrite formation by CLM, compound A or B and quercetin (standard) were calculated relative to the control. The IC50 values were calculated using the EZ-Fit enzyme kinetics software programme [11].

Statistical analysis
The results are presented as mean ± SEM. GraphPad Prism software was used to test differences between groups using one-way analysis of variance (ANOVA). p < 0.05 was considered significant.

Extraction
Sequential extraction of C. lancifolius with DCM and ME resulted in the collection of two extracts, which were labelled as CLD and CLM. The per cent yields of CLD and CLM were 1.33% and 2.11%, respectively.

Screening of secondary metabolites
Data of qualitative screening of CLM showed the presence of alkaloids, saponins, steroids and triterpenoids, flavonoids, cardiac glycosides, anthraquinones and tannins.

Physical and spectroscopic characteristics of 2,3,8-tri-O-methyl ellagic acid (compound A)
The spectra details for compound A were:  The proton NMR spectrum of compound A showed three methoxy group signals in the downfield region. Aromatic protons appeared at δ 6.36 (1H, singlet, H-8) with a disubstituted (ABX system) pattern. The signals at δ 6.95 (1H, d, with J = 4.2 Hz, H-5) and δ 11.47 showed hydroxyl group chelation at the H-5 position.
The broadband and depth ( 13 C-NMR) spectra of compound A gave 17 signals, data that indicate the presence of three methyl, two methine and 12 for quaternary carbons. The positions of three methoxy groups at C-3, C-2 and C-8 were determined based on 3 J HMBC correlations. All spectral, literature and physical data confirmed the identification of compound A as 2,3,8-tri-Omethylellagic acid. It was previously identified from the bark of Irvingia gabonensis [12]. See Table 1.

Physical and spectroscopic characteristics of 3-O-methylellagic acid 4-O-β D-glucopyranoside (compound B)
As stated in Table 3  The broadband and depth ( 13 C-NMR) spectrum of compound B revealed 21 carbon signals: one methylene, seven methine and 12 quaternary carbon signals.  [13].
The structures of compounds A and B were assigned from 1 H-and 13 C-NMR spectra in combination with electrospray ionisation mass spectrometry (ESI-MS) experiments and comparison with literature data of related compounds ( Table 2). Table 3 shows the effect of CLD and CLM extracts and isolated compounds (A and B) treatment on different cancer and normal cell lines. CLM had more cytotoxic potential compared with CLD. Compound B showed greater cytotoxic effects compared with compound A. Table 4 shows the antioxidant potential of bioactive CLM and isolated compounds in two in vitro assays. CLM showed better free radical scavenging compared with the isolated compounds. Compound B showed a better free radical scavenging effect compared to compound A.

DISCUSSION
The data presented in the current study are a continuation of a previous research report [8] in which it was found that extracts prepared using different parts of C. lancifolius contain compounds with antioxidant and cytotoxic potentials. However, the previous study reported antioxidant and cytotoxic potentials of crude extracts. Therefore, the current study was conducted to isolate compounds from C. lancifolius whole plant extracts following bioactivity-guided isolation methods.    It is a well-known fact that the majority of anticancer agents available on the market today are of a natural origin [14]. However, the available cytotoxic agents have mostly limited selectivity towards cancer cells. Hence, researchers around the globe are in a continuous search to identify and isolate novel molecules that have more specific actions towards cancer cells [15]. The current study was designed to isolate bioactive compounds from the most active whole C. lancifolius plant extracts. Data from the previous study showed profound antidiabetic activity of methanolic extract in alloxan-induced diabetic rabbits [5] and antioxidant activity [8].
Free radicals are chemical compounds with unpaired electrons; they are formed either endogenously (as a waste product of nutrient metabolism) or exogenously (tobacco smoking, ionising radiation, air pollution, organic solvents and pesticides). An excessive amount of these reactive oxygen species (ROS) is harmful to the body because these agents have the potential to cause tissue injury and even cell death [16]. The damage caused by free radicals may lead to many chronic diseases, such as cardiovascular disorders, neural disorders, mild cognitive impairments, Parkinson's and Alzheimer's disease, ulcerative colitis, aging, atherosclerosis and cancer [17]. Medicinal plants are important sources of antioxidant compounds; these natural antioxidants reduce the risk of many chronic diseases [18]. The antioxidative effects of phytoconstituents are mainly due to their redox properties which play an important role in neutralisation of free radicals, quenching of singlet and triplet oxygen species and/or decomposition of peroxides [19]. It is known that the antioxidant effect of a compound is proportional to its number of hydroxyl groups [20]. In the current study, 2,3,8-tri-Omethylellagic acid, 3-O-methylellagic acid 4-O-β D-glucopyranoside and the CLM showed significant antioxidant properties, which are probably due to the high number of hydroxyl groups they contain.

CONCLUSION
The investigation of a methanol extract of Pakistani C. lancifolius (aerial part) yielded two compounds: 2,3,8-tri-O-methylellagic acid and 3-O-methylellagic acid 4-O-β-D-glucopyranoside. These compounds show promising anticancer and antioxidant potentials, which is in agreement with previous studies. The phytochemical and pharmacological potential of C. lancifolius have not yet been completely explored; however, the findings of this study are a confirmation of the plant's chemical constituents, biotic properties and potential application as previously reported.