Phenolics of Achillea fragrantissima growing in Egypt and its cytotoxic activity

Achillea fragrantissima Sch. Bip. (Asteraceae) has been reputed in folk medicine of the Arabia region and Egypt for the treatment of many diseases. Air dried aerial parts of A. fragrantissima were extracted with 70% aqueous ethanol and fractionated with petroleum ether, dichloromethane, n-butanol and ethyl acetate fractions. Among them, dichloromethane showed the highest activity and was subjected to separation and purification by various chromatographic techniques. Five flavones and one phenolic acid were isolated and identified as acerosin (1), cirsimaritin (2), cirsiliol (3), luteolin (4), apigenin (5) and caffeic acid (6), respectively. Acerosin, the flavone of higher concentration isolated for the first time in genus Achillea, showed promising cytotoxic and radical scavenging activity. Molecular modeling of acerosin indicated its ability to bind in a co-crystallized ligand-like manner with kinase enzyme (FAK), namely with Cys502, alongside extra binding interactions with Glu430, Lys454 and Asp564. Total protein profiling of A. fragrantissima DNA was performed using polyacrylamide gel electrophoresis (PAGE) for characterization of plant where the total number of bands recovered was 4. 
 
   
 
 Key words: Acerosin, hepatocellular carcinoma, molecular modeling, DPPH, kinase enzyme (FAK).

As one of the most important folk medicines in the Arabian region and in Egypt, several biological studies were performed, demonstrating anti-neuroinflammatory, strong antioxidant and hepatoprotective activities (Elmann et al., 2011a, b;Khaled et al., 2010). Remarkable antiviral activity was observed against poliomyelitis-1 virus (Soltan and Zaki, 2009), in addition to potent larvicidal (Sathiyamoorthy et al., 1997) and antidiabetic activities (Yaniv et al., 1987;Ezzat and Salama, 2014). Studying its anticancer activity, A. fragrantissima was revealed as potent inducer of differentiation, cell cycle arrest and apoptosis in chronic myeloid leukemia and T cell lymphoma (Alenad et al., 2013). Cirsiliol (5, 3ʹ, 4ʹ-trihydroxy-6,7-dimethoxyflavone) isolated from A. fragrantissima showed spasmolytic activity, as it antagonizes the spasmodic effects, inhibits Ca 2+ influx and stimulates Ca 2+ released from intracellular stores (Mustafa et al., 1992). Applications of biochemical genetic techniques have an important potential to provide a new tool for the study of wild species, as well as domesticated species. Electrophoresis methods are considered a rapid and accurate test to identify and characterize species. Total protein profiling obtained by polyacrylamide gel electrophoresis (PAGE), is frequently used in resolving systematic relationships (Paradies and Ohms, 1987;Lioi et al., 1999;Karihaloo et al., 2002).
A. fragrantissima, very rich in its phenolic content, still represents an attractive field of study. The aim of this study was to shed light on its importance as a cytotoxic agent against hepatocellular carcinoma of wide occurrence in Egypt, and isolation of its bioactive compounds and in silico trials in order to identify the mechanisms behind which the bioactive compounds elicit performed as a means of identification and characterization of this plant species growing in Sinai.

General experimental procedures
For column chromatography, microcrystalline cellulose (E. Merck, Darmstadt, Germany) and sephadex LH-20 (Pharmacia, Uppsala, Sweden) were used. For paper chromatography, Whatman no. 1 sheets and Whatman no. 3 sheets for screening and preparative paper chromatography (PPC) (Whatman Ltd, Maidstone, Kent, England) were used. The pure compounds were visualized under UV light (254 and 365 nm). Solvent systems S1 (n-BuOH/HOAc/H2O: 4:1:5 v/v/v top layer) and S2 (15% aqueous HOAc) were used. The nuclear magnetic resonance (NMR) spectra were recorded at 300 ( 1 H) and 125 ( 13 C) MHz on a Varian Mercury 300. The results were reported as δ (ppm) values relative to TMS in the convenient solvents. Electron ionization-mass spectrometry (EI-MS) analyses were run on a Shimadzu QP-2010 Plus mass spectrometer at 70 ev and 220°C ion source temperature. UV analyses for pure samples were recorded on MeOH solutions and with different diagnostic UV shift reagents on a Shimadzu UV spectrophotometer (1800 UV probe).

Sulforhodamine B (SRB) assay
Human hepatocellular cancer cell line, HePG2, was obtained from Vacsera (Giza, Egypt). The cytotoxicity of the total 70% aqueous ethanol extract (AEE) and its successive extracts were tested against liver (HePG2) cancer cell line by sulforhodamineB (SRB) assay, as described by Skehan et al. (1990) and Al-Abd et al. (2008) at concentrations of 1000, 100, 10, 1 and 0.1 µg/ml. The dose response curves of different fractions were analyzed using Emax model (Equation 1) and percentage of cell viability was calculated using the following equation: In this equation, R is the residual unaffected fraction (the resistance fraction), [D] is the drug concentration used, Kd is the dose of the drug that produces a 50% reduction of the maximum inhibition rate and m is a Hill-type coefficient. IC50 was defined as the drug concentration required to reduce fluorescence to 50% of that of the control (that is, Kd = IC50 when R = 0 and Emax = 100 -R). The concentration required to reduce cell viability by 50% (IC50) was determined using the sigmoid Emax model. The extract with the highest activity and lowest IC50 was explored for its major constituents.

Microtetrazoline (MTT) assay
The major constituent with higher concentration was exposed to cytotoxic investigation. Cell viability was assessed against HePG2, using SRB assay (as previously described) and MTT (3-(4, 5dimethylthiazol-2-yl) -2, 5-diphenyltetrazolium bromide) assay via mitochondrial dependent reduction of yellow MTT to purple formazan (Mosmann, 1983). Cells were incubated either alone (negative control) or with different concentrations of the tested samples to give final concentrations of 100, 50, 25, 12.5, 6.25, 3.125, and 1.56 µg/ml. A positive control (doxorubicin) was used as a known cytotoxic natural agent, which gives 100% lethality under the same conditions (Thabrew et al., 1997). The percentage of change in viability was calculated according to Equation 2: (2)

Evaluation of the antioxidant effect (in vitro)
The activity of 1, 1-diphenylpicrylhydrazil (DPPH) was estimated according to the method described by Shimada et al. (1992). Radical scavenging activity was measured at 517 nm (Equation 3).
Where: A0 is the absorbance of the control reaction (DPPH) and A1 is the absorbance of the test extract. Ascorbic acid was used as standard (Oktay et al., 2003).

Quantitative estimation of total phenolic and flavonoid contents
The total phenolic content (TPC) was determined through the Folin-Ciocalteau reagent (FCR), using gallic acid as standard (Sellappan and Akoh, 2002). TPC was expressed as milligram of gallic acid equivalent (GAE) per milligram of plant extract for each fraction, based on gallic acid calibration curve (standard curve equation: y = 0.0011x + 0.0009, R 2 = 0.9867), having the absorptivity maxima at 765 nm. Quantitative estimation of flavonoid contents was performed via aluminum chloride assay, using quercetin as standard. Flavonoid content (FC) was expressed as milligram of quercetin equivalent (QE) per milligram of extract for each fraction, based on quercetin calibration curve, having the absorptivity maxima at 415 nm (the standard curve equation: y = 0.005x -0.0198, R 2 = 0.9774) (Chang et al., 2002;Kosalec et al., 2004).

Docking experiment
The molecular model of compound 1 (acerosin, 6,8,4ʹtrimethoxyluteolin), the major constituent in the CH2Cl2 extract, was built in its dissociated form using standard bond lengths and angles, with the MOE software suite version 2008. Following geometry optimization, a Monte Carlo conformational search was carried out with the ConfSearch module implemented in MOE. All molecular mechanics computations were performed with the Merck Force Field (MMFF94s). The experimental crystallographic structure of the kinase enzyme FAK complex was retrieved from the Protein Data Bank (PDB ID 3PXK). Missing hydrogens were added to the enzyme and partial charges were calculated. After removing the cocrystallized inhibitor, docking of acerosin was carried out using MOE. The target protein was kept rigid while the ligand was left free to explore the conformational space inside the enzyme cavity; 200 separate docking simulations were run using default parameters.

Protein banding analysis and Electrophoresis
Total cellular proteins were extracted from the sample, followed by analysis with electrophoresis, staining and destaining for protein gel according to Laemmli (1970) and Stegmann et al. (1988). The gel was placed between two sheets of cellophane membrane, dried on a gel drier for 2 h and photographed. The banding profile and the molecular weight of each band were calculated by visual comparison to a molecular weight marker.

SRB assay
Cytotoxic activities of different extracts (Pet. ether, CH 2 Cl 2 , EtOAc, BuOH, AEE) of A. fragrantissima were assessed against HePG2 cell line using SRB assay (Table 1). With zero resistant fraction, CH 2 Cl 2 fraction exhibited the highest effect against liver cell carcinoma (IC 50 = 38.7 ± 2.13 µg/ml), compared with other extracts showing cytotoxic effects, but with less potency and with  resistant fractions ranging from 1.4 to 11.3%.

Evaluation of the antioxidant effect (in vitro)
Comparing the radical scavenging activities of the different extracts, IC 50 and IC 90 were evaluated and presented in Table 2. EtOAc showed the lowest IC 50 (5.4 ± 0.52 µg/ml), followed by dichloromethane, while the highest IC 50 (35.3 ± 1.86 µg/ml) with the lowest radical scavenging activity was observed with AEE.

Quantitative estimation of total phenolic and flavonoid contents
The TPC and TF of different extracts of A. fragrantissima were estimated colorimetrically. EtOAC showed the highest phenolic content (34.52 ± 0.63 mg/g GAE) using FCR while CH 2 Cl 2 showed the highest flavonoid content (29 ± 0.17 mg/g QE) using the aluminium chloride assay ( Table 2).

Acerosin bioactivity
Due to the close relationship between the cytotoxic and the antioxidant activities, radical scavenging activity of acerosin was estimated and compared to ascorbic acid (positive control) using DPPH assay. The percentage of radical scavenging activity of acerosin was estimated as 84.5% at 100 µg/ml compared to 100% for ascorbic acid and AEE at the used concentration. Cytotoxic activity against HePG2 was evaluated using SRB and MTT assays. SRB assay showed IC 50 of 24.69 ± 1.24 µg/ml while MTT assay showed IC 50 of 35.6 ± 2.1.

Docking study
A conformational search using an implicit solvent model was accomplished for acerosin; this was followed by refinement of the geometry of local minima through a quantum-mechanical (QM) method. Subsequently, flexible docking of the compound was performed in the crystallographic structure of the kinase enzyme (FAK) recovered from the Protein Data Bank (PDB ID 3PXK). The poses obtained for acerosin were found to bind in a co-crystallized ligand-like fashion with FAK, namely with Cys502, along with extra binding interactions with Glu430, Lys454 and Asp564 (Figures 1 and 2).

Seed protein profiling
The seed protein profile of A. fragrantissima species was studied and the observed protein banding profile is enlisted in Table 3, according to the molecular weight of the revealed bands. Protein band patterns were coded '0' or '1', depending on their absence or presence. The molecular weights observed ranged between 215 and 44 KDa. The total number of bands recorded was 4 ( Figure  3).

DISCUSSION
Cancer has created a challenge to the medical science researchers. Presently, cancer remains the second leading cause of death in the world, following cardiovascular diseases. In addition, cancer presents a rationally complex set of difficulties because of multiple sites and causation, inefficiently understood biology and numerous intervention approaches (Vicker et al., 2009).  Discovery of natural products with potential anticancer activity represents a key step in the development of anticancer therapy. Previous cytotoxic studies of different Achillea species revealed that flavonoids represent major cytotoxic agents. Santoflavone (tetramethyl flavone) isolated from A. talagonica Bioss. Is shown as a good cytotoxic agent using Brine shrimp method (Saeidnia et al., 2009). In addition, centaureidin (O-methylated flavonol) isolated from A. clavennae shows the highest cytotoxic activity, compared with other flavonoids and sesquiterpenes (Trifunovic et al., 2006). Enhancement of the lipophilicity of these molecules results in higher potency (Ghantous et al., 2009).
Liver cancer is a very serious and solid tumor, which is highly endemic with hepatitis viruses. Egypt has the highest prevalence of hepatitis C virus (HCV) in the world, estimated nationally as 14.7% (El-zanaty and Way, 2008). In this study, CH 2 Cl 2 fraction exhibited the highest effect against liver cell carcinoma with zero resistant fraction, therefore, prone to further isolation and purification procedures. Previous reports on A. alexandriregis Bornm. & Rudsky revealed that combined chloroform and ethyl acetate extracts exhibit a pronounced cytotoxic effect against HeLa cancer cells (IC 50 = 25.92 ± 4.96 µg/ml) (Kundaković et al., 2005). Radical scavenging seems to play a considerable role in the antioxidant activity. In recent years, flavonoids have attracted a tremendous interest as potent-free radical scavengers and possible therapeutics against free radical mediated diseases (Middleton et al., 2000;Amić et al., 2003).
In order to identify the active components in the most effective fraction, CH 2 Cl 2 extract was subjected to cellulose column and eluted fractions were grouped into four main fractions based on PC examination. Furthermore, purification using PPC and sephadex LH-20 resulted in the isolation of six compounds identified on the basis of spectral data (UV, NMR, EI-MS) as well as comparison with published data (Wollenweber et al., 2003;Wallage, 1971;Mesquita et al., 1986;Lin et al., 2006;Bai et al., 2011;Nagao et al., 2002;Marder et al., 1996), and identified as five flavones (1-5) and one phenolic acid (6). Acerosin (1) was identified for the first time in genus Achillea.
Due to the higher yield of acerosin in the active fraction, further biological studies were conducted. Several assays have been developed for testing the cytotoxic activity. The effectiveness of SRB assay is frequently compared to that of MTT assay. SRB assay relies on the ability of SRB to bind to protein components of cells that have been fixed to plate by trichloroacetic acid, where the amount of dye extracted from stained cells is directly proportional to mass (Vichai and Kirtikara, 2006). The MTT assay requires cellular metabolic activity to convert the colorless tetrazolium to the purple colored formazan dye; therefore, it detects only viable cells. However, SRB assay is preferred to MTT assay, as SRB staining is independent of cell metabolic activity; therefore, fewer steps are required to optimize assay conditions for specific cell lines in SRB assay than in MTT assay (Keepers et al., 1991). IC 50 tested with SRB assay was lower than MTT assay, in contrast to Rubinstein et al. (1990).
Due to the close relationship between the cytotoxic and the antioxidant activities, radical scavenging activity of acerosin was estimated and compared to ascorbic acid (positive control) using DPPH assay. The percentage of radical scavenging activity of acerosin was estimated as 84.5% at 100 µg/ml compared to 100% for ascorbic acid and AEE at the used concentration.
Lower antioxidant effect of acerosin, compared with AEE, may be mainly due to the methylation at 4ʹ position, as Rice-Evans et al. (1996) and Cao et al. (1997) studied the relationship between the antioxidant activities and structure of flavonoids and concluded that the hydroxyl groups in the C-3′ and C-4′ of the B-ring are essential for the antioxidant activities. On the other hand, synergistic effects between different isolated compounds suggested a higher radical scavenging activity for the crude extract. In an attempt to rationalize the cytotoxic activity profile exhibited by acerosin, a molecular modeling study was carried out. FAK, a member of the protein tyrosine kinases (PTK) family, was implicated in increased survival and invasiveness in human carcinoma. Acerosin was found to bind with FAK, namely with Cys502, along with extra binding interactions with Glu430, Lys454 and Asp564.
Luteolin's anticancer property was reported to be associated with the induction of apoptosis and inhibition of cell proliferation, metastasis and angiogenesis. Furthermore, luteolin sensitizes cancer cells to therapeutic-induced cytotoxicity by suppressing cell survival pathways (Lin et al., 2008). Luteolin and luteolin analogs act as potent FAK inhibitors, thus playing an important role in the reduction of cancer cell survival and migration (Huang et al., 2005). Seed protein profiling is one of the methods for characterizing different species. Further molecular studies are necessary for the identification of the genes encoding enzymes specific for flavone and flavone ether that are responsible for the activity in A. fragrantissima. Results of the present study have clearly demonstrated that aerial parts of A. fragrantissima possess pronounced cytotoxic activity. No reports related to the cytotoxic activity of A. fragrantissima extracts against hepatocellular carcinoma have been found so far. Besides this, this study is the first to identify acerosin in genus Achillea, showing its cytotoxic and radical scavenging activity.

Conclusion
Acerosin, isolated from dichloromethane extract in A. fragrantissima, displayed pronounced cytotoxic activity against liver cell carcinoma cell line, as well as high radical scavenging activity. This bioactive compound is a good candidate for possible semi-synthesis and derivatisation to enhance its activity as a lead compound in cancer therapy. Further researches would be needed for the exploration of its mechanism of action.