Secondary metabolites, including flavonoids and other phenolic compounds, were extracted from ground mature roots of S. baicalensis (common name Huang Qin) using 95% ethanol. Commercially available roots from a California importer of Chinese herbal medicines (Win Hop Fung, Los Angeles) were obtained and ground into a fine powder using a Scientific Apparatus™ soil mill. The root powder was stirred for 24 hours in a 95% ethanol solution in order to extract the secondary metabolites and flavonoids. The crude extract was filtered through a PTFE membrane filter using a Swinney (Millipore Corp., Bedford, MA) filtering device. The extract was dried and quantified for the total amount of crude extract. A stock solution was prepared at 25 mg of solid per ml in absolute ethanol and was further diluted with sterile water immediately before treatment of the cells to achieve a concentration of 10 mg/ml in a 40% (v/v) ethanol solution.

Cell lines grown from primary and recurrent GBMs from three patients (Table 1) were cultured using previously established protocols 39 . Primary tumors are designated with a 2-letter code (ME, DI) or the last 2 digits of the year followed by a 2-letter code (00WA). The recurrent tumors from the same patient receive the same code with the addition of an "R" (ME/MER, DI/DIR). Following surgery for their primary tumor, both patients from whom we received recurrent tumor samples received 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU, carmustine) and radiation therapy. Despite therapy, both tumors recurred and the patients underwent additional surgery. Tumor cell line 00WA (Table 1) was selected for use in this study as a low-passage cell line derived from a primary tumor and serial passaged 5 times. Normal glial cells (HJ) grown from a sample obtained from a crainiotomy done for treatment of trauma were also cultured following previously established protocols 39 . All cell lines were cultured in Waymouth's MAB 87/3 (Mediatech, Herndon, Virginia) containing 20% Fetal Bovine Serum (FBS; Intergen Co., Purchase, NY) and incubated at 37°C in a humidified atmosphere of 5% CO₂. To select for the subpopulation of cells resistant to 10 μg/ml BCNU, cells were treated with gradually increasing concentrations of drug as described 40 41 . Cells were washed with MAB without serum; they were then mock-treated using MAB alone or treated with increasing concentrations of BCNU in MAB (2.5, 5.0, 7.5, 10 μg/ml) for 1 hour at 37°C with 5% CO₂. Cells were subsequently washed and fed with MAB containing 20% FBS. The cells were treated for three consecutive days then allowed to grow. Treatment was repeated several times until the cells in culture were resistant to 10 μg/ml of BCNU (ME Drug/MER Drug and DI Drug/DIR Drug) as evidenced by the absence of cell death compared to that observed in the mock-treated cultures. Cells were re-treated with 10 μg/ml BCNU every 8–10 passages to maintain the resistant phenotype.

Cells were seeded at a density of approximately 1 – 4 × 10³ cells in 200 μl per well. Following a 24-hour incubation period, cells were treated with 100 μg/ml of S. baicalensis extract containing 0.4% ethanol (v/v) in 200 μl of medium. Mock treatment consisted of a 0.4% (v/v) solution of ethanol in 200 μl of medium. Cellular metabolic activity was assayed using AlamarBlue™ (Serotec, Raleigh, NC) and the manufacturer's protocol at 0, 1, 3, 7, 10, and 15 days following treatment. The fluorescent reading was performed using a microtiter plate reader (485 nm λ excitation, 595 λ nm emission).

A Colony Forming Assay (CFA) was used to verify that the cells remaining metabolically active following treatment with S. baicalensis extract were also capable of proliferation. Cells were diluted and divided into aliquots containing approximately 1.5 × 10³ cells. Aliquots were treated with a single dose of 0 (untreated), 10, 25, 50, 75, 100, 150 and 200 μg/ml of the extract in MAB 87/3 medium supplemented with 20% FBS for 1 hour. Following incubation, cells were washed three times with fresh medium. Cells were resuspended in the medium and aliquots of 1.5 ml containing approximately 2 × 10³ cells were dispensed into three 60 mm culture dishes. Cells were cultured for at least 6 divisions, approximately 14 days, before they were fixed with methanol and stained with 4% Giemsa stain. Colonies, defined as 50 or more cells, were counted and plotted as a percentage of the control (untreated) colonies.

The cells were plated at approximately 1 × 10⁵ cells per well in 6-well tissue culture plates in 2 ml of medium and incubated at 37°C at 5% CO₂. After 24 hours, the medium was removed and replaced with fresh medium plus 20% FBS and supplemented with S. baicalensis extract (0, 25, 50, 100, 150, and 200 μg/ml) or a combination of S. baicalensis extract and BCNU (2.5 or 5 μg/ml). Cells were harvested 48 hours after treatment by digestion with 0.25% trypsin-EDTA solution at 37°C for 2–3 minutes. The cells were stained with trypan blue and live cells were enumerated. Cell counts were expressed as mean ± standard deviation (SD).

All results are expressed as mean ± standard deviation (SD). Statistical differences between correlated samples were evaluated using Student's t-test and noted to be significantly different where p < 0.05. Student t-test calculations were assessed on the VassarStats Statistical Computation website 42 . Composite treatments were compared using one-way analysis of variances (ANOVA) and considered significantly different where probability values were found to be equal to or less than 0.05. All ANOVA tests, as well as mean and SD calculations, were performed using GraphPad Prism (GraphPad Software, Inc., San Diego, USA).

The metabolic inhibitory effects of an extract from S. baicalensis were studied in cell lines cultured from paired primary (ME) and recurrent (MER) glioblastoma multiforme (GBM) tumors obtained from a single patient, as well as primary and recurrent cells that have been selected in vitro for resistance to 10 μg/ml of BCNU (ME Drug/MER Drug). Metabolic activity was assessed at 0, 1, 3, 7, 10, and 15 days following treatment with 100 μg/ml of the extract using an AlamarBlue metabolic assay. All wells were subconfluent when the experiment was begun to allow for changes in cell growth. In this way, we would see a difference regardless of whether the effect of the extract was cytotoxic or cytostatic. There was a significant (p < .05; n = 5) difference in the metabolic activity exhibited by cells treated with S. baicalensis extract compared with mock- or ethanol treated cells. There was no increase in metabolic activity exhibited in the ME and MER glioma cell lines, as well as subpopulations of these cell lines that were selected for BCNU resistance (ME Drug and MER Drug) for the 15 day duration of the experiment (Figure 1), while the untreated glioma cells (controls) exhibited exponential increases in metabolic activity. As an additional control, glioma cells were treated with a 0.4% (v/v) ethanol solution (solvent vehicle). Although there was some difference in fluorescence between untreated and ethanol treated MER cells at day 7 (Figure 1D) and 10 (Figure 1C), there was good growth in all four cell lines and the ethanol-treated cells showed no significant decrease in metabolic activity compared to the untreated control cells by the fifteenth day of culture. Microscopic examination showed extensive cytopathic effect (CPE) in cells treated with the S. baicalensis extract (data not shown).

Colony forming assays were done to demonstrate a dose response to the S. baicalensis extract and to show that the effect is cytopathic and not cytostatic. Cells from the paired primary and recurrent tumors ME/MER and DI/DIR prior to and following selection for resistance to 10 μg/ml BCNU were treated with 0 – 200 μg/ml S. baicalensis extract as described in the methods section. Cells were allowed to grow to form colonies of at least 50 cells. As seen in Figure 2, cells from all 8 cell lines showed a dose-dependent response to the S. baicalensis extract.

To demonstrate that the effect of the S. baicalensis extract on glioma cells was not a result of long-term in vitro cultivation of the cells, we tested the extract on cell line 00WA after 5 serial passages in vitro. This was the minimum number of passages required to obtain enough cells to perform the assay. Cells were treated with varying concentrations of the S. baicalensis extract as described in the Methods section, and the percentage of viable cells was determined by trypan blue exclusion 48 hours later. Figure 3 shows that the extract caused a dose-dependent reduction in viability similar to that observed in the ME/MER and DI/DIR cell line series.

To assess whether the extract demonstrates toxicity in normal cells, normal human glial cells (HJ) were plated and allowed to grow until they were close to confluence. These cells were then cultured in the presence and absence of various concentrations of the extract for 48 hours. As shown in Figure 4, normal glial cells treated with 25–100 μg/ml of S. baicalensis extract for 48 hours displayed no significant change (p > .05; n = 3) in metabolic activity assayed using AlamarBlue and compared to the control (untreated) cells. For comparison, a recurrent glioma cell line (MER) was also treated with 25-100 μg/ml of the extract for 48 hours. The glioma cells treated with S. baicalensis extract at concentrations of 50 and 100 μg/ml displayed a significant reduction (p < .05, n = 3) in metabolic activity when compared to the metabolic activity of the control (untreated) cells.

Microscopic examination of the cells following treatment with 100 μg/ml of the extract showed marked differences in the normal glial cells compared to the tumor cells (Figure 4B). The MER culture demonstrated an increase in the number of detached cells floating in the medium. The floating cells exhibited atypical morphologies such as DNA condensation, a hallmark of cell death (Figure 4B). The normal glial (HJ) cells did not show a substantial increase in floating cells or cells with DNA condensation or other hallmarks of cell death.

To determine if the S. baicalensis extract could potentiate the effects of conventional chemotherapeutic agents, we tested the effect of combined treatment using 50 μg/ml of S. baicalensis extract and 2.5 μg/ml of BCNU on the viability of cells from primary and recurrent cells selected for resistance to 10 μg/ml BCNU (Figure 5A). As expected, 2.5 μg/ml of BCNU alone had no effect on these cells and 50 μg/ml of S. baicalensis extract reduced viability to 15–30% of control; however, when used in combination survival was reduced substantially, particularly in the DI and DIR cell lines. The differences between the individual treatments and the combined treatment were found to be significant in all but the ME drug cells by one-way ANOVA (ME Drug: P = 0.0557 F = 8.785; MER Drug: P = 0.0116 F = 27.77; DI Drug: P = 0.0005 F = 253.3 and DIR Drug: P = 0.0010 F = 144.1). Fewer than 5% of any of the cells survived when the BCNU dose was increased to 5 μg/ml, even though this dose alone caused little, if any, CPE (data not shown). Microscopic examination of the combined effects of BCNU and the S. baicalensis extract again indicated an increase in the number of detached (floating) cells with atypical morphologies when compared to cells treated with either BCNU or the extract alone (Figure 6).

The effect of combined treatment on low passage, BCNU naive 00WA cells showed a similar significant potentiation of the drug effect, although the overall effect was less than that observed with the ME and DI cell series (Figure 5B). BCNU or S. baicalensis extract alone reduced viability to 75%; however, combination treatment resulted in a reduction of viability to 30% of the control, mock-treated cells (p = 0.0141 F = 24.20 by one-way ANOVA).