Metabolomic profiling, antioxidant capacity and in vitro anticancer activity of some compositae plants growing in Saudi Arabia

The present study was conducted to evaluate the metabolic profiling, antioxidant capacity and anticancer activities of some common widely grown plants of the family Compositae. The total phenolics, flavonoids, anthocyanins, saponins, total antioxidant capacity (TAC) and 2,2-diphenyl1picrylhydrazyl (DPPH) assays were determined in the selected plant extracts. In vitro anticancer activity was also assessed using human hepatocellular carcinoma (HepG-2) and breast adenocarcinoma (MCF-7) cell lines. The plant species revealed different metabolomic profiling. Artemisia showed the highest contents of the detected secondary metabolites compared to other plant extracts. Pulcaria crispa showed the highest inhibition concentration 50% (IC50) among the screened extracts against HepG-2 (8.9 μg/ml) and MCF-7 (8.14 μg/ml). The high performance liquid chromatography analysis (HPLC) of P. crispa extract revealed the presence of high content of three phenolic compounds, benzoic, chlorogenic acid and vanillic acid, along with two polyphenolic compounds, hesperidin and quercetrin. In summary, among the screened extracts, P. crispa has the most potent anti-tumor activity in vitro against HepG-2 and MCF-7 cell lines.


INTRODUCTION
Kingdom of Saudi Arabia (KSA) flora is rich with wild medicinal plants. In traditional medicine, these plants are used to treat several human diseases (Alyemeni et al., 2010;Aboul-Enein et al., 2012;Kuete et al., 2013). Moreover, there are several studies that reported on the metabolic, antioxidant and anticancer activity of some wild growing medicinal plants in the Kingdom of Saudi Arabia (Sher et al., 2012). Recently, due to the severe side effect of using the conventional chemotherapy for cancer treatment, several studies have been carried out to find either new and safe targeted therapy against cancer cells only, or to find natural compounds that ameliorate the side effects of the conventional chemotherapy (Patra et al., 2002;Kumar and Harlev et al., 2012). Therefore, recent studies have been directed to screen, and evaluate the new compounds naturally present especially in the medicinal plants (Patra et al., 2002). For instance, the methanolic extracts of 40 species of plants traditionally used in KSA for the treatment of several diseases were tested for their anticancer activity (Almehdar et al., 2012). By using human breast cancer (MCF-7) and human leukemia (HL-60) tumor cell lines, it has reported that Sesbania grandiflora extract has potent in vitro cytotoxic activity (Ritesh et al., 2011).
The family Compositae is one of the largest families of flowering plants used in traditional medicine. For instance, Pulicaria crispa is used to treat inflammation, as an insect repellent, and as a herbal tea (Ross et al., 1997). Achillea fragrantissima is used to treat respiratory diseases and gastrointestinal disturbances (Elmann et al., 2011). Therefore, several studies have been done to investigate its anticancer activity. For example, by using human chronic myeloid leukemi (K562), T cell lymphoma (Jurkat) and hepatocellular carcinoma (HepG-2) cell lines, A. fragrantissima extract showed a potential anticancer activities in vitro (Alenad et al., 2013). In other study, the cytotoxic effect of an aqueous extract of A. fragrantissima against HepG-2 cell line was reported (Thoppil et al., 2013). A recent study showed that the anti-proliferative activity of the hydro-alcoholic A. fragrantissima and A. falcata extracts against the MCF-7 cells did not possess cytotoxic activity up to 200 μg/ml (Hammad et al., 2014). Artemisia campestris aerial parts exhibited growth inhibition of HT-29 human colon cancer cells (Akrout et al., 2011). Aerial parts of P. crispa showed cytotoxicity in EJ-38 human bladder carcinoma cell (Stavri et al., 2008). Interestingly, it has been reported that plants grown under desert stress conditions produced a high concentration of secondary metabolites that possess a wide range of pharmacological effects, including anticancer activities (Harlev et al., 2012). The aim of this study was to assess the metabolic profiling, antioxidant and anticancer activity of some plants that belong to family Compositae that grow widely in the northern region of KSA.

Collection of plant materials and preparation of plant extracts
Achillea falcata (leaves-flowers), A. fragrantissima (leavesflowers), Artemisia judaica (leaves-flowers and seeds), Artemisia sieberi (leaves-flowers and seeds), P. crispa (shoot system), Pulcaria incisa (shoot system) and Rhyanteruim eppaposum (shoot system and flowers) were collected from desert, around Sakaka City, Aljouf Region, KSA. The plant materials were identified and authenticated by taxonomist at Camel and Range Research Center, Sakaka, Aljouf, KSA. The material was shade dried then ground to powder using an electrical mortar. The powered material was stored in air tight container until further use. The shade dried (50 g) powder of each plant material was filled in a conical flask containing 80% methanol. After 4 days, the extract was filtered and  concentrated in a rotary evaporator at a temperature not exceeding 50°C.

Determination of total phenolics and flavonoids
Total concentration of phenolics in the extracts was determined using Folin-Ciocalteu reagent with gallic acid as a standard, and expressed (mg) as gallic acid equivalents per gram of extract, according to Singelton et al. (1999). Total flavonoids content was determined using the aluminium chloride colorimetric method with quercetin as a standard, and expressed (mg) as quercetin equivalent per gram of extract according to Zhishen et al. (1999).

Determination of saponins
Saponins content was determined using vanillin solution, according to Ebrahimzadeh and Niknam (1998), and expressed (mg) as saponins equivalents per gram of extract.

Determination of anthocyanins
The anthocyanins content of the plant was determined according to the modified method by Padmavati et al. (1997). 100 mg of plant materials was dissolved in acidified methanol in well closed tubes covered with aluminium foils and incubated in a refrigerator for 24 h. The absorbance was read at 530 nm and 657 nm. The concentration was calculated using the following equation: anthocyanin concentration (μmol/g) = ([A530 -0.33 x A657]/31.6) X (volume [ml]/weight [g]).

Determination of total antioxidant capacity and 2,2-diphenyl-1picrylhydrazyl (DPPH) assays
Total antioxidant capacity (TAC) was determined using phosphomolybednum method, according to Prieto and Pineda (1999). The antioxidant capacity was expressed as ascorbic acid equivalent. Free radical scavenging activity of the sample extracts was determined spectrophotometrically using the method of Blois (1958), after obtaining crude extracts from the samples through evaporation of the solvent. The scavenging activity on the DPPH radical was expressed as inhibition percentage using the following equation: % radical scavenging activity = (Ac -As/Ac) x 100, where Ac = Absorbance of negative control at 517 nm and As = Absorbance of sample at 517 nm (Wang and Mazza, 2002).

Plant extracts and doxorubicin preparation for in vitro use
Different concentrations of each methanolic extract were prepared at 6.25, 12.5, 25, 50, and 100 µg/ml dissolved in DMSO (1%). Doxorubicin (Dox.) was also prepared at the same concentrations, 6.25, 12.5, 25, 50, and 100 µg/ml dissolved in DMSO (1%) under the same conditions and used as a positive control.

Determination of inhibition concentration 50% (IC50) for extracts using sulforhodamine B (SRB) colorimetric assay
The cytotoxicity of the plants extracts was tested against MCF-7, HepG-2 cell lines by SRB assay, according to Vichai and Kirtikara (2006). Briefly, the adherent cells were collected after trypsinization using 0.25% Trypsin-EDTA then washed twice and plated in 96-well plates at 1000 to 2000 cells/well. Cells were exposed to different extracts for 72 h and subsequently fixed with 10% trichloroacetic acid (TCA) for 1 h at 4°C. After several washings using distilled water, cells were exposed to 0.4% SRB solution (dissolved in 1% glacial acetic acid) for 10 min in dark place. A 1% glacial acetic acid was used to wash the plates several times. After drying overnight, Tris-HCl was used to dissolve the SRB-stained cells and color intensity was measured at 570 nm with micro plate reader. The results were linear over a 20-fold range of cell numbers and the sensitivity is comparable to those of fluorometric methods.

Identification and quantification of phenolics and flavonoids by HPLC
Analyses were carried out using a Perkin-Elmer HPLC system (USA) equipped with a binary LC-250 gradient pump and LC-290 UV/Vis detector. Samples were separated on C18 Hypersil ODS column (100 x 4.6 mm) with 5 μm particle size according to Ruiz et al. (2011). The mobile phase consisted of eluent A, 3.0 % acetic acid in water (v/v) and eluent B, methanol. The elution gradient was: at 0 min, 0% B; at 10 min 10% B; at 40 min, 70% B; at 50 min 0% B at a constant flow rate of 1 ml min -1 . Phenolic compounds were monitored by absorption at 280 and 330 nm. All measurements were performed in triplicates. Individual phenolic compounds of each sample were identified by comparing their relative retention time with those of the standard mixture chromatogram. A mixture of 17 standard (HPLC grade) phenolic compounds was used for the HPLC analysis. The concentration of each identified compound was calculated by comparing its peak area with that of the comparable standard, then converted to mg phenolic g -1 dried extract. All standards and solvents were HPLC spectral grade.

Statistical analysis
One-way analysis of variance (ANOVA) through MINITAB (version 12.21) was used to test the significance of quantitative data of phytochemical analysis. For SRB assays, the Sigma Plot (version 12) program was used to analyze the generated data.

Metabolomic profiling and antioxidant capacity of plant extracts
The potentiality of anticarcinongenic activity of a plant extract depends on the metabolomic profiling of that extract. Metabolomic profiling of ten extracts of 4 genera belonging to family Compositae were detected through the determination of the TAC and DPPH radical scavenging activity. The results showed that each genus from Asteraceae has its specific metabolomic profiling.

Artemisia and Pulicaria plants showed the highest content of phenolics
As shown in Figure 1A, the phenolics content showed specific pattern of variation among different genus, different species and organs. The highest content of phenolics content was registered in Artemisia and Pulicaria plants.  and Pulicaria incisa (MPT-52) showed higher content of phenolics comparing to A. judaica (MPT-34, leaves and flowers) and P. crispa (MPT-32), respectively. Although there is no significant difference between the content of phenolics in both Artemisia species, there was a significant difference (p = 0.020) detected between P. incisa and P. crispa (MPT-52 and MPT-32). Significant difference was also detected between the content of phenolics in Achillea falcata and A. fragrantissima (MPT-31 and MPT-33).

Achillea, Artemisia and Pulicaria have a higher content of flavonoids than Rhyanterium
Three genera of Compositae (Achillea, Artemisia and Pulicaria) showed high levels of flavonoids, while the fourth (Rhyanterium) showed moderate level of flavonoids ( Figure 1B). Both of A. judaica and A. sieberi (MPT-34 and MPT-30) showed the highest content of flavonoids among other species and organs followed by A. falcata (MPT-31) and P. incisa (MPT-52). The content of flavonoids showed significant difference between A. falcata and A. fragrantissima (p = 0.002). Significant difference (p = 0.017 and 0.005) was also observed between the content of flavonoids in A. judaica, A. sieberi seeds (MPT-15 and MPT-51) and R. eppaposum (MPT-2), respectively.

Saponins content showed highly variation among different species and different organs
Saponins content in plant extracts showed highly variation among not only Compositae genera but also among different species and different organs of each species (Figure 2A). The genus Artemisia showed the highest content of saponins comparing to other genera followed by Achillea. Among species and plant parts of Artemisia, A. sieberi (MPT-30) showed the highest content of saponins. Seeds of A. judaica (MPT-15) showed higher content of saponins than seeds of A. sieberi (MPT-51) (in both cases no significant difference between the content of saponins in both A. sieberi and A. judaica leaves and flowers or seeds was detected). A. falcata showed higher content of saponins than A. fragrantissima (MPT-31 and MPT-33). P. incisa (MPT-52) showed higher content of saponins than P. crispa. In cases of Achillea and falcata, a significant difference (p = 0.002, 0.025) between the content of saponins in the two parts of the two genera was observed.

Artemisia and Rhyanterium showed the highest content of anthocyanins of all tested genera
Artemisia (MPT-15 and MPT-51) and Rhyanterium (MPT-17) showed the highest content of anthocyanin compare to other extracts but only significant difference was observed between R. eppaposum shoot and flowers (MPT-2 and MPT-17 ) ( Figure 2B).

TAC and DPPH radical scavenging activity
Regarding total antioxidant capacity, as shown in Figure  3 and
The results showed also that the IC50 (Table 2 and Figures 4 and 5).

The inhibition concentration 50% (IC50) of different extracts on MCF-7 cell lines
The inhibition concentration 50% (IC50) of the previous plant extracts was also determined using MCF-7 cell lines     Figures 6 and 7), respectively. The 5 most effective extracts of the evaluated plants against carcinogenic cell lines were sorted out, and the ANOVA revealed that the 5 extracts are significantly differed from each other in phenolics, flavonoids, saponins, TAC and DPPH radical scavenging activity. Pearson's correlation showed that the TAC is significantly positive correlated with the content of phenolics in the 5 plant extract (p = 0.026).

HPLC analysis for phenolic of P. crispa
Based on the phytochemical analysis and anticancer activities of the tested plant extracts, the P. crispa methanolic extract showed the highest cytotoxic effect against HepG-2 and MCF-7 cell lines in vitro. To this end, the study further analyzed this extract by HPLC to determine the major phenolic and flavonoid compounds. The results showed that the major phenolic compounds were benzoic acid, chlorogenic acid and vanillic acid  Figure 6). Moreover, two flavonoid compounds namely hesperidin and quercetrin were identified as a major content in the extract (Table 4 and Figure 8).

DISCUSSION
In this study, the 4 genera of family Compositae showed different pattern of variation regarding the secondary metabolites content. The results showed difference in metabolomic profiling among the different species of the same genera, and among different organs of the same species. The 4 genera under evaluation could be arranged as following, regarding their metabolic content: Artemisia > Pulicaria > Achillea > Rhyanterium. This finding is in agreement with Abdel-Farid et al. (2014). Also, consistent with this study finding, different species of Acacia showed different metabolomic profiling by spectrophotometric analysis by Abdel-Farid et al. (2014).
In this study, was found the extracts which have the highest secondary metabolites content such as those of A. sieberi, P. incisa and R. eppaposum which also showed the highest TAC and DPPH radical scavenging activity. This finding is in agreement with previous reports such as Basar et al. (2013) and Abdel-Farid et al. (2014), which showed that there is a significant positive correlation between various secondary metabolites content such as phenolics, flavonoids, saponins and anthocyanins with the total antioxidant capacity. Abdel-Farid et al. (2014) reported TAC was found to be positively correlated with saponins and flavonoids content. Based on this study data, some plant extracts were found to be rich with some important secondary metabolites which could contribute significantly to the anticancer activity.
In the present study, the cytotoxic effect of 10 alcoholic plant extracts from 4 genera of family Compositae was evaluated by SRB assay using HepG-2 and MCF-7 cell lines in vitro. According to the American National Cancer     Institute (NCI), the criterion of cytotoxic activity for the crude extracts is an IC50 less than 30 ug/ml (Suffness and Pezzuto, 1990). Based on this fact, this study found that five out of ten extracts showed higher anticarcinogenic potentiality against HepG-2 and/or MCF-7 cell lines. P. crispa, P. incisa (shoot system), A. fragrantissima (leaves and flowers) and R. eppaposum (shoot system) showed higher cytotoxic activity against HepG-2 cell lines (less than 30 μg/ml). However, among the tested extracts, only P. crispa (shoot), R. eppaposum (shoot), A. sieberi (leaves and flowers) and P. incisa (shoot) showed higher cytotoxic effect against MCF-7 cell lines (less than 30 ug/ml). The strong anticarcinogenic activity of these extracts against HepG-2 and MCF-7 cell lines may be attributed to the high content of some secondary metabolites such phenolics, flavonoids and saponins in these extracts. The study finding is in agreement with Alenad et al. (2013). They found that A. fragrantissima extract showed a potential anticancer activity in vitro against HepG-2 cell lines. Furthermore, in other study, the cytotoxic effect of an aqueous extract of A. fragrantissima against HepG-2 cell line was reported by Thoppil et al. (2013). In a recent study by Hammad et al. (2014), they showed that the anti-proliferative activity of the aqueous and hydro-alcoholic A. fragrantissima and A. falcata extracts against the MCF-7 cells did not possess cytotoxic activity up to 200 μg/ml, and this finding is not consistent with this study. Consistent with this study findings, the aerial parts of P. crispa showed cytotoxicity in EJ-38 human bladder carcinoma cell (Stavri et al., 2008).