Lipofectamine was obtained from Invitrogen (Carlsbad, CA, USA). This liposomal formulation is composed of a 3:1 mixture of cationic lipid DOSPA and neutral lipid DOPE. Oleanolic acid, diosgenin, 2,2-diphenyl-1-picrylhydrazyl (DPPH), and Hoechst^® 33342 dye were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Thin layer chromatography silica gel plates (60F₂₅₄), ascorbic acid, hexane, ethanol, acetone, methanol, ethyl acetate, chloroform and dichloromethane were purchased from Merck KG (Darmstadt, Germany).

The roots of Asparagus racemosus were collected from Tak Province, Thailand. The plants were identified by the botanical staff, and the voucher specimen of the plant was deposited in the herbarium of the Faculty of Pharmaceutical Sciences, Naresuan University, Phitsanulok, Thailand.

Asparagus racemosus roots were dried and minced into powder, and then successively extracted with solvents starting from non-polar (hexane) to more polar (water) solvents. The hexane fraction was filtered and evaporated by a rotary evaporator (Eyela^® A-3S; Tokyo Rikakikai, Japan), after which soft yellow paste was obtained and named AR1-1. The marc was further extracted with 95% ethanol, filtrated and evaporated, after which the dark brown viscous liquid was obtained and named AR1-2. The residual marc was further extracted with water at 80°C. After filtration, the aqueous fraction was divided into two portions, one of which was evaporated and freeze-dried until brown powder was obtained and named AR1-3. The other portion was added to acetone and the resulting precipitate was separated out from the solution. Both the solution and precipitate were evaporated, freeze-dried, and named AR1-4 and AR1-5, respectively.

TLC is one of several techniques useful for the identification of phytochemical compounds (19,20). In this study, solutions of oleanolic acid and diosgenin (0.5 mg/ml) in methanol were used as known references for triterpene and steroid structure, respectively (21,22). The samples of the Asparagus racemosus extracts were prepared by dissolving fractions of AR1-1, AR1-2 and AR1-3 to AR1-5 (10 mg/ml) in ethyl acetate, methanol and water, respectively. All samples were filtered through a 0.2-μm nylon membrane filter, after which 8 μl was spotted by Autospot equipped with a scanner and camera (Camag^®; Limonat V, Switzerland) on TLC silica gel plates and air-dried for 5 min. The plates with the oleanolic acid and AR1-1 samples were developed with system I (2:1, hexane:ethyl acetate) (19), while those of diosgenin and AR1-2 to AR1-5 samples were developed with system II (6.4:5:1, chloroform:methanol:water) (23). After completion, the plates were dried at 100–110°C for 5 min, sprayed with 10 ml anisaldehyde-sulfuric acid reagent, heated at 105°C for 5–10 min and visualized under white and UV light (366 nm) for TLC fingerprints. This experiment was performed in duplicate.

The total saponin content of Asparagus racemosus extracts was determined as described by Makkar et al (21) and Xi et al (24). Sample solutions of diosgenin (0.5 mg/ml) in methanol and those of AR1-1, AR1-2 and AR1-3 to AR1-5 (4 mg/ml) in dichloromethane, methanol and water, respectively, were prepared and filtered through a 0.2-μm nylon membrane filter before use. A 250-μl portion of 8% w/v vanillin solution in absolute ethanol was added to 250 μl of sample, which was subsequently mixed with 2.5 ml 72% v/v sulfuric acid, incubated in a water bath at 60°C for 10 min and finally cooled in an ice bath for 5 min. The absorbance of this mixture was measured at 544 nm (Perkin Elmer Lambda 35; Perkin Elmer, Waltham, MA, USA). The calibration curve of diosgenin at a high degree of linearity with a correlation coefficient (r²) of 0.9991 was achieved across the specified range of 8.3–41.6 μg/ml. The diosgenin equivalent (DGE) (μg/mg extract) was used as reference in determining the total amount of saponins in each fraction of the extracts. This experiment was performed in triplicate.

The DPPH assay for assessing the antioxidant activity of the Asparagus racemosus extracts was carried out as described by Brand-Williams et al (25), with minor modifications. The absorbance of the DPPH solution in methanol was measured at 515 nm (Perkin Elmer Lambda 35), and a calibration curve at a very high degree of linearity with a correlation coefficient (r²) of 0.9999 was achieved across the specified range of 10–100 μM. Solutions of AR1-1, AR1-2 and AR1-3 to AR1-5 (4 mg/ml) were prepared in dichloromethane, methanol and water, respectively, 200 μl of which was mixed with 1,400 μl of 60 μM DPPH solution. The decrease in absorbance was determined at 0, 1 and every 15 min until 60 min. The percentage of DPPH remaining at 60 min was determined and plotted as a function of the AR1-4 concentration. L-ascorbic acid was used as a positive control. The antioxidant activity was expressed in terms of the mean effective concentration (EC₅₀), which was defined as the amount of antioxidant that could decrease the initial DPPH concentration by 50%.

Human lung epithelial NCI-H460 cells were obtained from the American Type Culture Collection (Manassas, VA, USA). The cells were cultured in RPMI-1640 medium (Invitrogen) supplemented with 5% fetal bovine serum, 2 mM L-glutamine, 100 U/ml of penicillin and 100 μg/ml of streptomycin. The cells were grown in a humidified atmosphere of 5% CO₂ at 37°C until they reached over 80% confluence. The cells were subsequently passaged by the use of a solution containing 0.05% trypsin and 0.5 mM EDTA.

The aqueous fraction AR1-4 (10 mg/ml) was dissolved in water and filtered through a 0.2-μm nylon membrane filter before use. Cells (3×10⁴/well) were seeded in a 96-well plate in 100 μl of culture medium and grown for 24 h. The subconfluent monolayer of cells was washed once with PBS and once with RPMI-1640. Subsequently, 100 μl of RPMI-1640 was added, and the cells were untreated or pre-treated with various concentrations of AR1-4 (100, 500 and 1,000 μg/ml) for 1 h, and then treated with Lipofectamine (20 μg/ml) for 6 h. The untreated cells were used as a control. Apoptosis was determined by the Hoechst 33342 assay.

For Hoechst 33342 staining, cells were incubated with 10 μg/ml of Hoechst 33342 dye at 37°C for 30 min. Analysis for apoptosis was carried out by scoring the percentage of cells having intensely condensed chromatin and/or fragmented nuclei by fluorescence microscopy (Axiovert 25 CFI; Carl Zeiss, USA). Approximately 1,000 nuclei from random fields were analyzed for each sample. The percentage of apoptosis was calculated as (apoptotic nuclei/total nuclei) × 100%. The experiment was performed in triplicate.

Asparagus racemosus extracts contain saponins as the main constituents (26). The major constituents of Asparagus racemosus in methanolic extract are steroidal saponins, which have been separated, purified and termed ‘Shatavarin’. Ten derivatives of Shatavarin have been defined as Shatavarin I to X (1,4). Gautam et al (23) found that an Asparagus racemosus aqueous decoction contained steroidal saponin, alkaloids, proteins, starch, tannin and mucilage. Agrawal et al (7) carried out screening tests and reported positive results for steroids, phytosterols, carbohydrates, tannins, anthraquinones, saponins, glycosides and flavonoids, and negative results for terpenoids, amino acids and alkaloids. Most recently, Visavadiya and Narasimhacharya (26) found that most of the phytoconstituents in Asparagus racemosus root are saponins (8.83%), while the remaining constituents are polyphenols (1.69%), phytosterols (0.79%), ascorbic acid (0.76%) and flavonoids (0.48%).

Since saponins are found in both aqueous and methanolic extracts of Asparagus racemosus and can be purified successfully to pure compounds, they were considered as markers in this study. Standardization of Asparagus racemosus in this study involved the confirmation of the presence of saponins and control of the quality of Asparagus racemosus extracts by examining the TLC fingerprints and total saponin content.

The anisaldehyde-sulfuric acid reagent is generally used to detect phenol, sugar, steroid and terpene, which will turn violet, blue, red, grey or green, respectively. It was found that the TLC fingerprints of oleanolic acid and AR1-1 developed with system I exhibited a blue color, while those of diosgenin and AR1-2 to AR1-5 developed with system II exhibited yellow-green and green colors under white light. Detailed information regarding the R_f values and color zones is listed in Table I, and indicates that the compounds detected in AR1-1 to AR1-5 using this reagent may contain the structure of phenol, sugar, steroid and/or terpene in their molecules. Oleanolic acid, as triterpene, showed only a blue stripe, whereas AR1-1 exhibited many stripes with a blue color, one being on top of the lane. This indicates that AR1-1 may contain some saponins, including triterpenoid saponin. The more polar developing system II was applied to diosgenin and to fractions AR1-2 to AR1-5, which appeared to contain some steroidal saponins since green stripes similar to that of diosgenin appeared.

Among all of the fractions of the Asparagus racemosus extracts, the compounds of AR1-1 differed from those of AR1-2 to AR1-5. Fractions AR1-2 to AR-5 consisted of similar constituents as evidenced by the same pattern of TLC fingerprints with three main bands and a zone near the starting spot. Moreover, the band at an R_f value of 0.45 was prominent in AR1-2, while a zone near the starting spot was more pronounced in fractions AR1-3 to AR1-5. As a result, solubility of the former was increased using 95% ethanol as an extracting solvent, whereas solubility of the latter was increased using water as an extracting solvent. This was determined by the successive extraction using non-polar to more polar solvents.

Vanillin-sulfuric acid reagent was used for derivatization of the saponins having an OH group at their C-3 position to give chromogens. The chromogens formed were not dependent on the nature of the sugar moieties (21). Among all of the Asparagus racemosus extracts, the amount of DGE could be ranked in order as: AR1-2 ∼ AR1-3 ∼ AR1-4 > AR1-5 > AR1-1, as shown in Fig. 1. It is apparent that the highest amount of saponins was found in the 95% ethanol and aqueous fractions, but with a difference in the type of saponins.

In this study, ascorbic acid was used as a positive control, and AR1-4 was chosen as a representative based on its highest saponin content. Firstly, the reaction kinetics in the antioxidant activity of both ascorbic acid and AR1-4 with DPPH were carried out to determine the suitable incubation time before measurement. These were found to reach a steady state by 1 and 60 min, respectively (Fig. 2). The time to reach steady state, as classified by Brand-William et al (25), for reaction kinetics of ascorbic acid was found to be much more rapid (<1 min) than that of AR1-4 (60 min). Active compounds having antioxidant activity in AR1-4 would consist` of a large complex molecule with slow reaction kinetics, while small molecule of ascorbic acid could easily be subject to interaction with DPPH.

The mean effective concentrations (EC₅₀) of ascorbic acid and AR1-4, as shown in Fig. 3, were found to be 1.5 and 600 μg/ml, respectively. Recently, Potduang et al (27) showed that the EC₅₀ of a crude ethanol extract of Asparagus racemosus dry powdered root was 381.91 μg/ml, which was relatively lower than that of the aqueous fraction AR1-4 from successive extraction in this study. The antioxidant activities in terms of the % DPPH remaining of the various fractions of Asparagus racemosus extracts at a fixed concentration of 500 μg/ml are compared in Fig. 4; AR1-2 (95% ethanol fraction) significantly exhibited the highest antioxidant activity compared to the other fractions. These results were consistent with those reported above. In addition, the antioxidant activities of fractions other than AR1-2 were almost in the same order of magnitude.

Although Asparagus racemosus extracts exhibit far less antioxidant activity than ascorbic acid when tested in vitro with the DPPH method, numerous studies have reported their substantial antioxidant effect when tested in vivo (2,3,6,26). According to these studies, antioxidant enzymes (superoxide dismutase, catalase) and ascorbic acid were increased by the effects of Asparagus racemosus aqueous and methanolic extracts, although the underlying mechanism has not yet been elucidated. To date, there are no reports indicating that saponins of Asparagus racemosus extracts are the active compound responsible for this antioxidant effect.

The AR1-4 aqueous fraction was selected for the preliminary test to assess the antioxidant effect in cells, since its main constituents shown in the TLC fingerprint were similar to those of fractions AR1-3 and AR1-5 in terms of both main bands and intensity. Additionally, AR1-4 was more readily dissolved in water compared to the other fractions, which was more appropriate for the treatment of cells due to the non-toxicity of the water solvent. The results of the Hoechst assay showed that treatment of the H460 cells with Lipofectamine (20 μg/ml) caused extensive apoptosis compared to the control (Fig. 5). Upon pre-treatment of the cells with AR1-4, Lipofectamine-induced apoptosis was significantly decreased with an increasing concentration of AR1-4.

Toxicity and apoptosis induced by cationic liposomes as well as their mechanisms have been found to be clearly associated with ROS generation, which could effectively be inhibited by ROS scavengers or antioxidant enzymes, particularly superoxide dismutase (11,16,28). The protective effect of AR1-4 against Lipofectamine-induced apoptosis in this study can probably be attributed to the reduction in oxidative stress-induced toxicity imposed by its antioxidant activity. However, it is interesting to note that the antioxidant activity of AR1-4 with DPPH in this study was 400 times less than that of ascorbic acid, while it effectively protected the cells from apoptosis as compared to the untreated cells (Fig. 5). The results suggest that this antiapoptotic effect may be caused by the indirect enhancement of antioxidant enzymes, such as superoxide dismutase or catalase, rather than a direct antioxidant effect against ROS in the cells. This is supported by several studies showing that root extracts of Asparagus racemosus are capable of markedly increasing and protecting the inactivation of superoxide dismutase and catalase in rats (2,3,6,7).

In conclusion, Asparagus racemosus extracts appear to be composed of saponins, the contents of which are richer in extracts from polar compared to non-polar solvents. Successive extraction resulted in fractions having different antioxidant activities. In addition, aqueous fraction AR1-4 was shown to exhibit a significant protective effect against Lipofectamine-induced apoptosis in human lung epithelial H460 cells.