Antioxidant activity and in vitro inhibition of tumor cell growth by leaf extracts from the carob tree (Ceratonia siliqua)

Abstract

The methanol leaf extracts of female cultivars of the carob tree [Ceratonia siliqua L. (Fabaceae)] and of hermaphrodite and male trees were investigated for their contents of phenolic compounds, their in vitro antioxidant activity, measured by 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical-scavenging and linoleic acid system assays, and their in vitro tumor growth inhibition on HeLa cells. The different cultivars and trees showed high levels of phenols, and considerable variations in the amount of these compounds. The extracts showed significant radical scavenging activity (RSA), which was not significantly affected by the gender of the tree. From the female cultivars tested, Galhosa exhibited the highest RSA. Gender significantly affected the antioxidant activity of the extracts measured by the linoleic acid system assay, and males and hermaphrodites showed the highest activities. The extracts displayed a remarkable ability to inhibit tumor cell proliferation, and their bioactivity varied with different cultivars or trees tested. Extracts from male and hermaphrodite trees exhibited higher capacity to inhibit the proliferation of HeLa cells than the female cultivars.

Keywords::

• Cytotoxicity
• flavonoids
• HeLa cells
• polyphenols
• reactive oxygen species

Introduction

Free radicals (reactive oxygen species, ROS, and reactive nitrogen species, RNS) are products of normal cellular metabolism, and are extremely reactive and potentially damaging transient chemical species. The delicate balance between beneficial and harmful effects of free radicals is very important to living organisms and is maintained by “redox regulation” defense mechanisms, which include preventive and repair mechanisms, physical defenses, and antioxidant defenses provided by vitamins, carotenoids, sterols, and phenols (Valko et al., 2007).

There have been numerous studies on the biological activities of phenols, which are potent antioxidants and free radical scavengers (Sugihara et al., 1999). Furthermore, many isolated phenol compounds or extracts rich in these compounds have been reported to possess anticancer, anti-carcinogenic, anti-mutagenic, antibacterial, antiviral, or anti-inflammatory activities (Yamagishi et al., 2000; Greenspan et al., 2005; Lambert et al., 2005; Moreno et al., 2006; Song et al., 2007). Phenolic compounds may also have antioxidant effects used for managing oxidation stress-related chronic diseases such as diabetes and hypertension (Kwon et al., 2008).

The carob tree [Ceratonia siliqua L. (Fabaceae)] is one of the most useful trees of the Mediterranean basin. It is mainly used in the food industry as a source of gum extracted from the seeds (LBG; E410). The pulp is used in infants for the treatment of diarrhea of bacterial and viral origin (Loeb et al., 1989). The leaves and pulps have antioxidant (Kumazawa et al., 2002; Makris & Kefalas, 2004, Owen et al., 2003; Papagiannopoulos et al., 2004), antiproliferative (Corsi et al., 2002), and antimicrobial activities (Kivçak et al., 2002). Leaf extracts have potential anxiolytic and sedative effects (Avallone et al., 2002).

The chemical composition of pods varies among the different cultivars, and includes a high amount of different types of polyphenolic compounds (Avallone et al., 1997; Corsi et al., 2002; Kumazawa et al., 2002; Owen et al., 2003; Papagiannopoulos et al., 2004), which are also present in leaves in higher amounts (Corsi et al., 2002). Some previously isolated constituents include gallic acid, hydrolyzable and condensed tannins, flavonolglycosides, flavonoids (Owen et al., 2003; Papagiannopoulos et al., 2004), and pinitol (Baumgartner et al., 1986).

As far as our literature survey could ascertain, there is no information about the phenolic profile or biological activities of leaf extracts of Portuguese female cultivars of this species; nor are there any comparative studies of the variation in chemical composition between the three genders. In this context, we assessed the amounts of phenol compounds from the methanol leaf extracts of six representative Portuguese female cultivars of the carob tree growing in the Algarve region, and the leaf extracts of two male and two hermaphrodite individuals. The antioxidant activity of the extracts was assessed using two established in vitro methods, and this activity was compared to the activity of pure phenolic compounds. Furthermore, we investigated the antitumor activity of the extracts using HeLa cells, a cell line of human uterine cervix adenocarcinoma, as a model system.

Materials and methods

Collection of plant material and preparation of extracts

Samples from six of the most representative female Portuguese cultivars of the carob tree in terms of fruit productivity, namely Mulata, Galhosa, Aida, Costela/Canela, Gasparinha, and Preta de Lagos, and from two hermaphrodite (H1 and H2) and two male trees (M1 and M2) were sampled during August and September of 2005. The different cultivars and trees are from a cultivar collection from the Ministry of Agriculture, Rural Development and Fisheries (Direcção Regional de Agricultura do Algarve, Delegação de Tavira), and were identified by J. Graça. Fully expanded leaves were randomly collected from the middle third of the branches in all canopy orientations, dried at 40°C for 2 days, crushed and milled in a laboratory-scale hammer mill, and stored in the dark at −20°C until extraction. The extracts were prepared as described by Owen et al. (2003). Samples (10 g) were extracted in a Soxhlet apparatus first with n-hexane (2 × 3 h), to remove lipids, and then with methanol (5 h). The methanol extracts were centrifuged at room temperature (RT) during 10 min at 3000 rpm, the supernatant was collected, and the solvent was removed by rotary evaporation with vacuum. The extracts were resuspended in methanol and kept at −20°C in the dark until analysis. Every extraction was performed at least twice.

Chemicals used for experimentation

Dulbecco’s modified Eagle medium (DMEM), l-glutamine, penicillin, and streptomycin were purchased from Biological Industries; fetal bovine serum (FBS) was from PAA Laboratories and n-hexane was from Labscan. Sigma Chemical Co. supplied DPPH (1,1-diphenyl-2-picrylhydrazyl), chlorogenic acid, (+)-catechin, linoleic acid, chloroform and β-carotene, whereas WST-1 [2-(4-iodophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium], monosodium salt) was purchased from Roche. Folin–Ciocalteu reagent, (–)-epicatechin, gallic acid, methanol (MeOH), Tween 40, p-dimethylaminocinnamaldehyde (DMACA), aluminum chloride (AlCl₃), and sodium carbonate (Na₂CO₃) were from Fluka Biochemika. Catechol was purchased from Acros organics. All the other chemicals where of analytical grade.

Measurement of total phenol content

The total phenol content of the extracts was determined in samples (2 mg/mL) by the Folin–Ciocalteu colorimetric method (Julkunen-Tiitto, 1985). Briefly, aliquots of the extracts (0.1 mL, 2 mg/mL) were added to 5 mL of distilled water and 0.5 mL of Folin–Ciocalteu reagent and vigorously shaken. After 3 min, 1 mL of a saturated solution of sodium carbonate (Na₂CO₃) was added, and the volume made up to 10 mL with distilled water. The mixtures were allowed to stand for 60 min at RT for complete reaction, and the total phenols were determined by colorimetry at 720 nm (Shimadzu UV-160A). The amount of total polyphenols was calculated as a gallic acid equivalent from the calibration curve of gallic acid standard solutions, and expressed as gallic acid equivalents (GAE) in milligrams per gram of initial dry plant material.

Determination of total condensed tannins content

The amount of condensed tannins was determined by the p-dimethylaminocinnamaldehyde (DMACA) method, according to Arnous et al. (2001). Samples (0.4 mL, 2 mg/mL) were mixed with 2 mL of DMACA solution (0.1% in 1 N HCl in MeOH). The mixture was vortexed, allowed to react at RT for 10 min, and the absorbances were read at 640 nm on a Shimadzu UV-160A spectophotometer. The concentration of condensed tannins was estimated from a calibration curve, constructed by plotting known solutions of (+)-catechin, and results were expressed as catechin equivalents (CE) in milligrams per gram of initial dry plant material.

Determination of total flavonoids

The flavonoid content was estimated in the extracts by the aluminum chloride (AlCl₃) method (Lamaison & Carnat, 1990). In short, 1 mL of methanol extract (2 mg/mL) was added to 1 mL of 2% methanol AlCl₃.6H₂O. The absorbance was measured 10 min later at 430 nm (Shimadzu UV-160A). The results were expressed as rutin equivalents (RE) in milligrams per gram of initial dry plant material by comparison with the values obtained from standard rutin treated in the same conditions.

Determination of antioxidant activity

DPPH radical scavenging activity

The effect of extracts on the DPPH radical was estimated using the method of Brand-Williams et al. (1995). Aliquots (0.1 mL) of test sample at concentrations ranging from 400 to 50 μg/mL were reacted with 2.5 mL of methanol DPPH solution (100 μM) for 30 min in the dark, and the decrease in absorbance at 517 nm was measured spectrophotometrically (Shimadzu UV-160A). The DPPH radical scavenging activity (RSA) was calculated by the following equation: RSA (%) = [(control absorbance − sample absorbance)]/(control absorbance) × 100. Results were expressed as IC₅₀ value (μg/mL), which is the amount of sample necessary to decrease by 50% the absorbance of DPPH, and as antiradical efficiency (AE), which is 1000-fold the inverse IC₅₀ value (Parejo et al., 2003).

Linoleic acid system

The activity of the extracts was determined according to Tepe et al. (2006). A stock solution of β-carotene–linoleic acid mixture was prepared by dissolving β-carotene (0.5 mg) in chloroform (50 mL), and adding linoleic acid (25 μL) and Tween 40 (200 mg). Chloroform was completely evaporated using a vacuum evaporator, and 100 mL of distilled water was added with vigorous shaking. Aliquots (3 mL) of this reaction mixture were dispensed to test tubes, the extracts (0.1 mL) were added, and the emulsion system was incubated for 60 min at 50°C. Oxidation of the emulsion was assessed spectrophotometrically (Shimadzu UV-160A) by measuring absorbance at 470 nm after the 60 min period. The antioxidant activity is expressed as percent inhibition relative to the control after a 60 min incubation using the following equation: AA = 100(DRC – DRS)/DRC, where AA is the antioxidant activity, DRC is the degradation rate of the control [= ln(a/b)/60], DRS is the degradation rate in the presence of the sample [= ln(a/b)/60], a is the initial absorbance at time 0, and b is the absorbance at 60 min. Extracts and polyphenol compounds were evaluated at the final concentrations of 1 mg/mL in the assay mixture. Both the scavenging and the antioxidant activities of the extracts were compared to those of pure phenolic compounds, namely gallic acid, (+)-catechin, chlorogenic acid, (–)- epicatechin, and catechol.

Antitumor activity

In vitro bioassay test

Human cervical adenocarcinoma cells (HeLa line) from the American type culture collection (Rockville, MD) were grown and maintained in Dulbecco’s modified Eagle medium (DMEM) containing 1 g/mL of glucose, and supplemented with 10% (v/v) fetal bovine serum (FBS), l-glutamine (2 mM), penicillin (50 U/mL), and streptomycin (50 μg/mL). The cells were grown in an incubator at 37°C, in a 5.1% CO₂ humidified atmosphere.

Tumor growth inhibition was determined by the WST-1 assay (Choudhary et al., 2006; Wang et al., 2006). Exponentially growing cells were seeded on 96-well plates at a density of 7 × 10³ cells/well. After incubation for 24 h at 37°C, cells were treated with several dilutions of the extracts (400–25 μg/mL), and incubated for 24, 48, and 72 h. Two hours before the end of incubation, 20 μL of WST-1 were added to each well and further incubated for 2 h at 37°C, and the optical density (OD) was measured at 450 nm on a microplate reader (Power Wave XS spectrophotometer). The cytotoxicity of the extracts was expressed as the percentage viability of the cells, half maximal inhibitory concentration (IC₅₀), and maximal degree of inhibition.

Statistical analysis

The experimental results are expressed as mean ± standard deviation (SD). All experiments were conducted in triplicate, and all tests and measurements were repeated at least three times. The data were analyzed using the analysis of variance (ANOVA) method to assess differences with the SPSS statistical package for Windows (release 15.0; SPSS Inc.), and significance between means was tested by Duncan’s new multiple range test (p = 0.05). The correlation between antioxidant and free-radical scavenging activities versus the polyphenol, tannin, and flavonoid contents was determined using the correlation and regression program of Microsoft Excel. IC₅₀ values were calculated by sigmoidal fitting of the data in the GraphPad Prism V 4.0 microcomputer program.

Results

Polyphenol contents

The total content of phenol compounds was not different between extracts from trees of different genders (p > 0.05). Among female trees, the total amount of phenols varied among different cultivars; the levels of these compounds were markedly higher in the female cultivars Galhosa, Aida, and Preta de Lagos than in the cultivars Mulata and Gasparinha (Table 1).

Table 1.  Contents of total phenol compounds, condensed tannins and total flavonoids in methanol leaf extracts of male, female and hermaphrodite carob trees.

Cultivar/tree	Total phenols (mg GE g^−1 DW)	Condensed tannin (mg CE g^−1 DW)	Flavonoids (mg RE g^−1 DW)
Female
Mulata	16.4 ± 1.5 c	16.7 ± 0.1e	2.1 ± 0f
Galhosa	39.4 ± 2.3 a	23.7 ± 1a	13.4 ± 0.4a
Aida	34.2 ± 7.4 a	17.8 ± 0.1c	7.0 ± 1.0c
Gasparinha	24.0 ± 0.5b	17.7 ± 0.7c	7.2 ± 0.3c
Costela/Canela	29.8 ± 3.2ab	19.9 ± 0.1b	6.9 ± 0.2c
Preta de Lagos	34.1 ± 3.1a	1.8 ± 0.3f	7.7 ± 0.5b
Hermaphrodite
H1	31.6 ± 2.7a	21.07 ± 0.2a	5.1 ± 0e
H2	29.7 ± 1.1ab	19.5 ± 0.8b	6.3 ± 0.2d
Male
M1	28.7 ± 2.4ab	16.1 ± 0.3d	6.0 ± 0.5d
M2	28.9 ± 2.8ab	16.3 ± 0.2d	7.9 ± 0.1b

Values represent means ± SD of 3 assessments. For each column and gender, statistical analysis was made between cultivars or trees. Values followed by different letters are significantly different at P < 0.05 (one-way ANOVA, Duncan’s New Multiple Range Test).

The amount of condensed tannins varied significantly between the different genders and cultivars (p < 0.05), ranging from 1.8 to 23.7 mg CE g⁻¹, with the lowest amount obtained from extracts from the female cultivar Preta de Lagos and the highest amount obtained from the female cultivar Galhosa and hermaphrodite tree H1 (Table 1). We found that the leaves of the hermaphrodites were richer in tannins, and the level of these compounds was markedly higher in tree H1. In female cultivars, tannins were present in highest amounts in the cultivar cultivar Galhosa (Table 1). While there was no significant difference in total flavonoid contents of extracts from different genders of tree, significant differences were observed between female cultivars (p < 0.05) (Table 1). Among the female cultivars, Galhosa and Mulata had the highest and lowest values, respectively, while among hermaphrodites, flavonoids were present in the highest amounts in leaves from tree H2 (Table 1). Among males, M2 had the highest content of these compounds (Table 1).

Radical scavenging activity (RSA)

The extracts showed a significant RSA (Table 2), which was concentration-dependent (data not shown). There was no gender-based variation in RSA (p > 0.05), but there was a significant variation in RSA between extracts taken from different cultivars and different individual trees (p < 0.001). Among the female cultivars, Galhosa exhibited the highest activity, denoted by the lowest IC₅₀ and highest antiradical efficiency values, while among the male and hermaphrodite trees, M1 and H2 exhibited the highest RSA, respectively (Table 2). Furthermore, the RSAs of extracts from the three genders were similar to those displayed by chlorogenic acid and (+)-catechin.

Table 2.  DPPH radical scavenging activity of methanol leaf extracts from female, male and hermaphrodite carob trees, and pure polyphenolic compounds.

	IC50 (μg/mL^−1)	AE
Female
Mulata	> 400	nd
Galhosa	94.6 ± 0.7	10.5 ± 0.1
Aida	217.3 ± 4.6	4.6 ± 0.1
Gasparinha	273.2 ± 11.9	3.6 ± 0.1
Costela/Canela	199.1 ± 5.5	5.0 ± 0.1
Preta de Lagos	175.7 ± 6.5	5.6 ± 0.2
Hermaphrodite
H1	167.2 ± 8.1	5.9 ± 0.3
H2	159.8 ± 18.7	6.3 ± 0.8
Male
M1	153.2 ± 17.0	6.5 ± 0.7
M2	159.6 ± 49.7	6.7 ± 2.3
Phenolics
Gallic acid	84.3 ± 15.6	12.0 ± 2.2
Chlorogenic acid	143.2 ± 2.5	6.9 ± 0.1
(+)-catechin	158.9 ± 15.6	6.3 ± 0.6
(−)-epicatechin	288.5 ± 28.4	3.4 ± 0.3
Catechol	49.1 ± 7.8	20.7 ± 3.5

Values represent means ± SD of 3 measurements of IC₅₀ and antiradical efficiency (AE).

A correlation analysis was conducted between the RSA and the amount of phenols, tannins, and flavonoids in all of the trees studied, and in the female cultivars (Table 3). For all trees studied, the RSA was closely correlated to the total amount of phenols and flavonoids, but was not correlated to the amount of tannins in the extracts. In extracts from female trees, the amount of total phenols and of flavonoids was very closely correlated with the RSA.

Table 3.  Correlation coefficients (R) between radical scavenging activity (RSA), antioxidant activity (AA) and contents of total phenols (GAE), tannins (CE) and flavonoids (RE).

Plant Material	N	R
		RSA/GAE	RSA/RE	RSA/CE	AA/GAE	AA/RE	AA/CE
Females	18	0.88^a	0.95^b	0.15	0.53	0.81^a	0.19
All the trees mixed	22	0.85^b	0.71^a	0.29	0.49	0.47	0.36

Note: GAE- gallic acid equivalents; CE- catechin equivalents; RE- rutin equivalents.

^a p<0.05; ^b p<0.01

Antioxidant activity in the linoleic acid system

Significant gender-based differences in the antioxidant activity of the extracts were observed (p < 0.05), with males and hermaphrodites having the highest activities (Figure 1). Overall, extracts from males and hermaphrodites are better than or as good at preventing the discoloration of β-carotene as pure phenolic compounds. As for the RSA, a correlation analysis was conducted between the AA and the amount of phenols, tannins, and flavonoids in extracts of all trees studied, and in extracts from the female cultivars (Table 3). For extracts from females, high correlation coefficients were found between the AA and the total amount of flavonoids, but not between the AA and the total amount of phenols or tannins. When all samples were considered, there was no correlation between the analyzed parameters.

Figure 1.  Antioxidant activity of methanol leaf extracts from female, male, and hermaphrodite carob trees, and of pure polyphenol compounds. Values represent means ± SD of three assessments of the percent inhibition of autoxidation of the linoleic acid/β-carotene emulsion. Extracts and phenol compounds were evaluated at final concentrations of 1 mg/mL. Values followed by different letters are significantly different at p < 0.05 (one-way ANOVA, Duncan’s new multiple range test).

Antitumor activity

Except for the female cultivar Galhosa, whose extracts had no effect on HeLa cell proliferation, carob extracts significantly inhibited HeLa cell proliferation in a dose-dependent manner (p < 0.05) (Figure 2). The arrest of HeLa cell growth was not time-dependent and was already evident after a treatment lasting 24 h. For the highest concentration tested (400 μg/mL), cell survival was generally below 25% (Figure 2). There was significant variation in bioactivity between the extracts from different genders of tree. Regardless of the concentration, treatment with extracts from males and hermaphrodites resulted in a significantly higher decrease in cell proliferation than treatment with extracts from females (p < 0.01). Extracts from the hermaphrodite tree H1 produced the highest inhibition capacity (IC₅₀ < 25 μg/mL), while extracts from the female cultivar Galhosa were the least effective at preventing the proliferation of HeLa cells (IC₅₀ > 400 μg/mL) (Table 4). The maximal inhibition values were high (generally 100%) (Table 4). A correlation analysis was performed between cytotoxic activity and RSA for all the trees studied and for the female cultivars, but no relationship was observed (data not shown).

Figure 2.  Effect of treatment with methanol leaf extracts from female (A), hermaphrodite (B), and male (C) carob trees on cell proliferation. Each point represents the mean of three independent experiments.

Table 4.  Inhibitory concentrations (IC₅₀) and maximal degree of inhibition (Max.) of methanol leaf extracts of female, hermaphrodite and male carob trees on HeLa cells.

Cultivar/tree	IC50 (μg/mL^−1)	Max. (%)
Female
Mulata	79 ± 3.7	100
Galhosa	> 400	–
Aida	150.3 ± 21.6	96.9 ± 0.7
Gasparinha	82.8 ± 8.9	100
Costela/Canela	67.01 ± 15.2	100
Preta de Lagos	143.82 ± 54.5	79.8 ± 16.9
Hermaphrodite
H1	149.9 ± 70.3	95.3 ± 1.5
H2	< 25	100
Male
M1	81.6 ± 5	100
M2	50.6 ± 0.5	100

Discussion and conclusions

The idea of using leaves of carob, a typical Mediterranean tree, as a source of polyphenols, was raised after the evidence that carob leaves contain a high amount of these compounds (Corsi et al., 2002). Our results support these findings since in this study, except for female cv. Mulata, all the cultivars and trees displayed remarkably high levels of total phenolic content, with GAE values >20 mg/g dry weight (DW) (Kähkönen et al., 1999). On the other hand, the amount of total phenols in leaf extracts from the carob tree were higher than those observed in Salvia spinosa L. (Lamiaceae), a plant species that is well known as a rich source of polyphenols (Alali et al., 2007).

Cross-varietal screening tests have repeatedly shown that certain genotypes within a plant species can have widely divergent levels of inherent antioxidants (Lila, 2006). In this work, we observed a general variation in the amount of phenolic compounds between extracts from different carob tree genders and different cultivars (Table 1). Working with the same species, Avallone et al. (1997) observed different levels of total polyphenols in different organs of female trees, and among trees from different locations. Emmons and Peterson (2001) also found significant differences among cultivars in concentrations of the majority of the phenol compounds measured in Avena sativa L. (Poaceae). Similar results were found by other authors in different plant species (Cetkovic et al., 2004; Maksimović et al., 2005; Gattuso et al., 2007; Li et al., 2007).

In this study, the free radical scavenging activity (RSA) of leaf extracts was assessed against DPPH radicals, and their antioxidant activity (AA) was assessed using the β-carotene–linoleic acid system. The extracts showed a significant RSA (Table 2), suggesting that they are capable of scavenging free radicals, and thus, may be able to prevent the initiation of free radical-mediated chain reactions by preventing the abstraction of hydrogen from susceptible polyunsaturated fatty acids. On the other hand the RSA varied between extracts from different genotypes. There was a significant gender-based difference in the AA of the extracts, and overall, extracts from males and hermaphrodites were equally or better able to prevent the discoloration of β-carotene than pure phenolic compounds (Figure 1).

Extracts with a higher phenol or flavonoid content generally show higher antioxidant activity, and good correlations have been found among these parameters (Alali et al., 2007; Li et al., 2007). Generally, the RSA was closely correlated to the total amount of phenols and flavonoids suggesting that these compounds are major contributors to the RSA of the extracts (Table 3). These results agree with previous reports which found that phenols and flavonoids contribute significantly to the RSA of different plant species (Parejo et al., 2003; Alali et al., 2007).

Interestingly, despite the fact that the extracts have a higher content of tannins than flavonoids, no significant correlation was found between RSA and the amount of tannins in an extract (Table 3), suggesting that flavonoids may be the major contributor to the RSA of the extracts. Flavonoids do indeed show strong antioxidant and radical scavenging activities (Pietta, 2000), and the use of flavonoid-containing drugs is associated with a reduced risk for certain chronic diseases, some cardiovascular disorders, and certain types of cancerous processes (Kris-Etherton et al., 2002). Their antiradical properties stem from targeting ROS that are implicated in the initiation of lipid peroxidation. Moreover, flavonoids are soluble chain-breaking inhibitors of the peroxidation process. They scavenge intermediate peroxyl and alkoxyl radicals and chelate metal ions, the latter of which are among the major components in the initiation of radical reactions (Pietta, 2000).

For female cultivars, high correlation coefficients were found between the AA and the total flavonoid content, but not with total phenol or tannin content (Table 3). When we considered all samples together, there was no correlation between the analyzed parameters (Table 3). These results are consistent with the findings of other authors in the same species (Alali et al., 2007), and suggest that apart from phenols and flavonoids, other compounds in the extracts may also contribute significantly to the antioxidant activity of carob tree leaf extracts. Moreover, a plant with high antioxidant activity, for which phenolic content versus antioxidant activity falls above the regression line, should be a plant in which novel antioxidants may be found, and the carob tree is such a plant (Alali et al., 2007).

The search for new therapeutic drugs for the treatment of cancer from natural products is based mainly on the cytotoxic properties of the natural samples (Goldin et al., 1981). In this sense, the incubation of HeLa cells with carob leaf extracts resulted in a remarkable concentration-dependent decrease of cell proliferation. Such an effect has been previously reported in several cancer cell cultures, after the application of different plant extracts (Wedge et al., 2001; Chen et al., 2004; Ramos et al., 2005). Cytotoxicity and the inhibition of cell proliferation are also displayed by several anticancer drugs derived from natural materials that are in wide clinical use, such as vincristine and vinblastine from Catharanthus roseus G. Don (Apocynaceae), paclitaxel (Taxol®) and taxotere from species of yew (Taxus), etoposide derived from lignans of Podophyllum spp. (Berberidaceae), and camptothecin analogs, such as topotecan, from Camptotheca acuminate D. (Nyssaceae) (Houghton et al., 2007).

Interestingly, the arrest of HeLa cell growth after treatment was not time-dependent and was already evident after a treatment lasting 24 h. This indicates that the polyphenols in the extracts promptly initiated a series of cellular events leading to the inhibition of cell proliferation and/or the induction of cell death. However, it should be noted that by using the WST-1 assay it is not possible to differentiate between cell growth inhibition and an increase in cell death. In this regard, data from studies focusing on elucidation of the molecular basis of the putative anticancer activity of polyphenols indicate that both the inhibition of cell growth and the induction of cell death play a role in the antitumor activity of polyphenols (Wenzel et al., 2000; Lazzè et al., 2004).

The bioactivity of the extracts varied between different cultivars and trees tested (Figure 2). Similarly, the extracts from male and hermaphrodite trees showed a higher antitumor activity, which also corresponds to the AA results. The RSA of the different extracts was directly correlated to the total amount of phenols and flavonoids, but there was no correlation relationship between RSA and antitumor activity (data not shown). These results suggest that the inhibition of tumor cell proliferation in vitro by the methanol leaf extracts of carob cannot be solely explained by the concentration of phenol/flavonoid compounds. Other phytochemicals may play a significant role in the antiproliferative activity.

This is the first time that the carob tree has been investigated for antitumor activity against human cancer cells. Taken together with results for the phenolic profile and antioxidant activities, our results suggest that the methanol leaf extracts possess exploitable antioxidants, and that they could be a source of phenolic compounds with potential antitumor activity. Our findings therefore serve as a prelude to in vitro and in vivo studies that may lead to verification of the efficacy of the antitumor activities of carob leaf extracts against various human cancers, elucidation of the mechanisms of their cytotoxic activity, and identification of the active components.

Declaration of interest: This work was partially supported by CICYT (CTQ2006-03794/BQU), Instituto de Salud Carlos III (CB06_01_0074 and PI060624), the Generalitat de Catalunya (2005SGR 00662), the Institute for Research in Biomedicine, and the Barcelona Science Park. One of the authors (L.C.) thanks the Portuguese Foundation for Science and Technology (FCT) for a post-doctoral grant (grant SFRH/BPD/20736/2004).