MATERIALS AND METHODS

Collection of medicinal plants: The dried plant material was purchased from Beijing Tong Ren Tang Chinese Herbal Medicine shop, Sydney, Australia. The scientific names and family names are given in Table 1. The plant materials were ground to a fine powder in a grinder before extraction.

Preparation of the water extract: Approximately 3 g of each grounded plant material was taken and extracted with hot water at 121°C for 1 h. The extracted samples were centrifuged at 10,447xg for 20 min, the supernatant was transferred into a 50 mL volumetric flask adjusted the volume to 50 mL.

Preparation of the ethanol extract : Ground samples (3 g) were extracted with 95% ethanol on a hot water bath set at 70°C for 6 h. The extracted samples were centrifuged and the supernatant was transferred into a 50 mL volumetric flask, then the volume was adjusted to 50 mL with 95% ethanol. The samples were stored at-4°C until analysis. All water and ethanol extracts were filtered before analysis.

GSK-3β inhibition assay: Based on Kinase-Glo system and his capacity for detect the ATP present after the enzyme reaction, an in vitro assay was developed to evaluate the inhibition of human GSK-3β. The kinase reaction consumes ATP, but the unused ATP is used by the Luciferase for catalyzed the reaction of transformation of Bettle-Luciferin in Oxyluciferin and light. This light could be measured. The enzyme is unable to react in presence of an inhibitor and the light generated by the Luciferase is increases due to unconsumed ATP (Rinnab et al., 2008).

Human recombinant GSK-3β was purchased from Millipore (Millipore Iberica S.A.U.) The pre-phosphorylated polypeptide substrate GS-1 (RRRPASVPPSPSLSRHS (pS)HQRR) was synthesized by American Peptide Company (Sunnyvale, CA). Kinase-Glo Luminescent Kinase Assay was obtained from Promega (Promega Biotech Ibérica, SL). ATP and all other reagents were from Sigma-Aldrich (St. Louis, MO). The Assay buffer contained 50 mM HEPES (pH 7.5), 1 mM EDTA, 1 mM EGTA and 15 mM magnesium acetate. Enzyme Buffer contained the same formulation as assay buffer in addition to TWEEN 20 0.001% which was included to increase the stability of the enzyme.

The assays were performed in black 384-well plates. First 10 μL of test extract and 10 μL of enzyme (10 ng) dissolved in Enzyme Buffer were added to each well and mixed. After 5 min 10 μL of substrate solution was added, containing 25 μM of GS-1 substrate and 1 μM ATP. After 30 min incubation at 30°C, 30 μL of Kinase-Glo Luminescent reagent was added to each well and the luminescence was recorded using a Victor2^TM Wallac multimode reader.

The inhibition percentage of GSK-3 β was determined by the equation:

Assay buffer with 1% DMSO was used as negative control and Alsterpaullone 500 nM was used as positive control. Alsterpaullone and thiadiazolidinones (TDZD), both known inhibitors of the enzyme, were used as internal controls.

Determination of cytotoxic properties
Cell lines: Fa2N4, an immortalized Hepatocyte cell line, were maintained in MFE Essential Support Medium F with MFE Culture Medium Supplement A. Collagen I flasks and plates were used to grow Fa2N4 cell line. All cell cultures were kept at 37°C under a humidified atmosphere of 5% CO₂.

Cytotoxicity assay: The Methylthiazolyldiphenyl-tetrazolium bromide (MTT) assay test was applied to five cell lines for evaluation of the cytotoxic activity. It is based on the ability of drug-treated cells to reduce the yellow water soluble substrate MTT into a dark blue formazan product that is insoluble in water. Nicotinamide adenine dinucleotide (NADH) is provided directly by the cells which in turn require proper metabolic function. Therefore, the MTT reduction rate is an indicator of the functional integrity of the mitochondria and, hence, of cellular viability. For the in vitro cytotoxic activity assay, the numbers of cells per culture wells were 100,000 on 96-well plates. Samples were incubated with 60-80% of confluent culture of each cell line for 24 hours in an atmosphere of 5% CO₂ at 37°C. Each extract was tested in two different concentrations. Absorption (OD) at 570 nm was measured in a Victor2^TM Wallac spectrofluorometer.

The inhibition percentage against the tested cell lines was determined by the equation:

Medium with 1% DMSO was used as negative control and medium with 2 mM of Methyl Methane Sulfonate (MMS) was used as positive control. Actinomycin D (250, 50, 12.5 μM), doxorubicin A (250, 50, 12.5 μM) and rotenone (250, 125 μM) were used as internal controls.

Data presentation and analysis: Data for the GSK-3 β screen and MTT cytotoxicity assay were analyzed using the gene data screener program (Genedata AG, Switzerland). DMSO at the same concentration used for the solubility of the compound, served as the negative control for assays, while 500 nM Alsterpaullone and 2 mM MMS were used as positive controls. In all results the RZ’ factor (Zhang et al., 1999) was between 0.7-0.9. The comparative analysis of various data sets was done using statistical correlation, was represented using Spotfire Silver (Tibco).

RESULTS
Traditional Chinese Medicinal (TCM) plants have been used for centuries as dietary supplements for symptoms of diabetes and neurological diseases. However, the in vitro effects of these extracts have not been determined. This study reports the GSK-3β inhibitory activities of water and ethanol extracts of 42 TCM plants. The inhibitory activities are expressed in terms of IC₅₀ values (μg mL^-1) of both aqueous and ethanol extracts. These results from all the selected plants are presented in Table 2 and Fig. 1 and 2. It is interesting to note that significant relationship was observed between the GSK-3β inhibitory properties and their traditional anti-diabetic and neuroprotective activities (Table 1). Significantly larger number of water extracts (33%, Fig. 1a) showed inhibitory activity without toxicity compared to the number of ethanol extracts (21% Fig. 2a) that showed activity.

Amongst the water extracts, 14 plants showed significant inhibitory activity (qIC₅₀<20 μg mL^-1) with no or minimal cytotoxicity (Table 2 and Fig. 1). Most active plants are: P. vulgaris (qIC₅₀<10.3 μg mL^-1 and non cytotoxic), R. rubescens (qIC₅₀<2.58 μg mL^-1 and moderately cytotoxic), Sanguisorba officinalis (qIC₅₀<2.58 μg mL^-1 and cytotoxic) and S. glabre (qIC₅₀<2.58 μg mL^-1 and non-cytotoxic). Several other plant extracts exhibited significant inhibitory activity (Table 2 and Fig. 1) indicating their high therapeutic index.

However, it should be noted that some of the water extracts (25%, Fig. 1b) were active but toxic and several other plant extracts (42%, Fig. 1c) did not show any activity.

In the case of ethanol extracts, only S. officinalis (qIC₅₀<2.58 μg mL^-1 and non-cytotoxic) showed significant inhibitory activity. Approximately, one third of the ethanol extracts (34%, Fig. 2b) showed inhibitory activity with toxicity. It is interesting to note that the inhibitory activities and toxicities are significantly different for water and ethanol extracts (Table 2, Fig. 1 and 2).

In order to further understand the GSK-3β inhibitory activities of the selected medicinal herbs in terms of their antioxidant contents (total phenolics and flavonoids), 11 water extracts and 10 ethanol extracts were selected (Table 3 and 4) and correlation plots were developed (Fig. 3 and 4). Amongst the water extracts the GSK-3β inhibitory activity showed significant correlation with total phenolics content (Fig. 3a, R² = 0.5146, p<0.05) and also with the total flavonoid content (Fig. 3b, R² = 0.5529, p<0.05). The correlation of GSK-3β inhibitory activities to their total phenolics content (Fig. 4a, R² = 0.7651, p<0.05) and flavonoid content (Fig. 4b, R² = 0.5384, p<0.05) was also significant in ethanol extracts.

The results presented in this study are in agreement with the fact that the total phenolics and flavonoid contents are contributors to the GSK-3β inhibitory activity of herbal medicine.

The water extracts of the herbs A. vulgaris, D. indica, L. lucidum and S. baicalensis showed high GSK-3β inhibitory activity and also have high total phenolics and flavonoid content (Table 2, 3 and Fig. 3). The ethanol extracts of the herbs S. officinalis, T. chinensis, S. suberectus and A. arguta showed high GSK-3β inhibitory activity and also have high total phenolics and flavonoid content (Table 2, 4 and Fig. 4).

DISCUSSION

Many plants, that showed significant activity with water extracts, have displayed minimal or no activity with ethanol extracts. A diagrammatic representation of the inhibitory activities of the plants and their cytotoxic properties are given in Fig. 1 (water extracts) and Fig. 2 (ethanol extracts). The results of this study clearly indicate that the w ater extracts display superior GSK-3β inhibitory activity when compared with ethanol extracts. Also the results indicate that water extracts are less toxic than the ethanol extracts and safer to use.

Following important conclusions can be drawn from Fig. 1 and 2:

Modern drug discovery is often inspired from the traditional knowledge of medicinal plants. In this regard, TCM plants have received huge interest due to their long history of usage in the treatment of various disorders. Traditionally, the TCM plants were consumed primarily in the form of hot water extraction or alcohol extraction (Parekh et al., 2005; Ravipati et al., 2012). One of the critical steps in the biological screening of medicinal plants is the type of extraction used, as a matter of fact each extraction method yields different active ingredients (Parekh et al., 2005; Ravipati et al., 2012). Water and ethanol extraction methods are both cost effective, easy to prepare the plant material and they are non-toxic at minimal dosages. The current study on GSK-3β inhibition potential of plant material was therefore conducted using both water and ethanol extracts. The cytotoxic properties of all the plant extracts were also evaluated as described in a previous study (Ravipati et al., 2013).

It has been reported in the literature that antioxidants play a crucial role in delineating the diabetic complications (Rahimi et al., 2005; Tchinda et al., 2008). In addition, studies have also shown a positive correlation between antioxidant content and α-glucosidase inhibition suggesting the role of antioxidants in the regulation of diabetic conditions (Apostolidis and Lee, 2010). For instance boswellic acid, ellagic acid, quercetin, rutin and normoglycemic are flavonoids that showed significant hypoglycemic and anti-diabetic activity in rats with STZ-nicotinamide induced type 2 diabetes. After 14 days of administration of STZ-nicotinamide in rats, the total cholesterol, triglyceride was significantly diminished, suggesting the anti-diabetic activities of flavonoids (Apostolidis and Lee, 2010). In agreement with previous studies, the GSK-3β inhibitory activities of the selected plants observed in this study are significantly correlated with their antioxidant potential (Ravipati et al., 2012; Ravipati et al., 2013). Tussilago farfara, Salvia miltiorrhiza and Paeouis suffuticosa contain large quantities of antioxidants and trace elements with significant antioxidant and anti-inflammatory activities (Ravipati et al., 2012; Ravipati et al., 2013). It is also observed that these plants exerted their maximum inhibitory activity against GSK-3β.

It is conceivable that modern drug discovery is inspired from the ethno-pharmacological evidence. Traditionally, the extracts of TCM plants were used in the treatment of diabetes and the whole plant extract was used in the treatment of epilepsy, irritability, insomnia and anxiety disorders (Ravipati et al., 2012; Ravipati et al., 2013). Enzyme assay guided fractionation studies carried out by (Gao et al., 2008) revealed the significant inhibitory activity of methanol extract of T. farfara showed highest inhibition against maltase (Apostolidis and Lee, 2010). Further characterization of this extract revealed the presence of 3, 4-dicaffeoylquinic acid (Parekh et al., 2005), 3, 5-dicaffeoylquinic acid (Ravipati et al., 2012) and 4,5-dicaffeoylquinic acid. An investigation carried out by (Huang et al., 2012) showed high therapeutic potential on diabetic conditions if the plants possess high total polyphenolic content (Huang et al., 2012). The rat models showed a significant decrease in the blood glucose, total cholesterol, triglyceride and blood urea nitrogen and increase in insulin sensitivity index, suggesting the anti-diabetic properties of the plant extracts (Huang et al., 2012). Another plant P. suffruticosa has been used in anti-diabetic herbal formulations (Shin et al., 2012) in order to evaluated their anti-diabetic properties in vitro models. These studies showed significant anti-diabetic affect by inhibiting the uptake of glucose (Lau et al., 2007). Studies (Ha et al., 2009) also suggest that the antioxidant content of medicinal plants/herbs may play a role in anti-diabetic properties. Remarkable fact to be noted is that the studies involving in vitro inhibitory activities gave most valuable supporting evidence for their traditional use as anti-diabetics and for other diseases like cancer, Alzheimr’s and inflammation.

CONCLUSION

Traditional Chinese Medicinal plants have been the source of many pharmaceutical compounds that are available in the market today. Current study is the first of its kind to screen large number of plant extracts for their GSK-3β inhibitory activities. Of all the plants studied, P. vulgaris, R. rubescens and S. glabre, showed highest GSK-3β inhibition. The results presented in this study clearly indicate that water is the best extraction solvent for isolating the compounds with GSK-3β inhibitory activity from medicinal herbs. Significant correlation was found between the antioxidant content and anti-diabetic ability of the plants. Further investigations employing various separation techniques on these plants could lead to the discovery of promising inhibitors of GSK-3β. Currently, bioactivity guided fractionation and characterization of novel class of molecules from these plants are underway in our laboratory.

ACKNOWLEDGMENT

This study was funded by the University of Western Sydney and the Fundación MEDINA, a public-private partnership of Merck Sharp & Dohme de España S.A./Universidad de Granada/Junta de Andalucía.