Cytotoxic Activities of Salvia. of the Labiatae Family

ABSTRACT

Eight crude extracts of five Salvia. species were evaluated for cytotoxic activities against brine shrimps and four human cancer cell lines [human colon adenocarcinoma (HCA), HepG2, MCF-7, and human pancreatic carcinoma (HPC) along with a normal mouse cell line (areolar cells)] as a control. In the brine shrimp lethality test, all samples, except S. fruticosa. L. (Sifnos collection) and S. verbenaca. L. (Zante collection), were found to be highly active with ED₅₀ values less than 300 μg/ml. In the case of human cancer cell lines, S. fruticosa., collected from Kalymnos and Crete, were active against HCA cells with LC₅₀ = 60.4 and 40.1 μg/ml respectively. Interestingly, only one sample, S. fruticosa. collected from Kalymnos, was active against HepG2 cells with LC₅₀ = 68.1 μg/ml. In the case of MCF-7 cells, S. fruticosa. collected from three different locations (Kalymnos, Rhodos, and Crete) showed similar activity with LC₅₀ = 43.1, 41.1, and 42.3 µg/ml, respectively. All S. fruticosa. samples were found to be cytotoxic toward a normal mouse cell line when tested at 0.1 mg/ml. All the other samples had LC₅₀ values greater than 75 µg/ml,and were considered to be inactive.

Keywords:

• Brine shrimp
• cancer cell lines
• cytotoxic activities
• Salvia. spp
• Labiatae

Introduction

Since ancient times, plants belonging to the Labiatae family have been used for treating various diseases, including cancer, in many countries (Hartwell, 1969). This family consists of about 250 genera and 6700 species and is considered to be a good source of natural antioxidants. As a part of our research studies for novel antitumor agents from plants (Badisa et al., 2000; Chaudhuri et al., 2002), we have recently examined a number of plants belonging to this family for cytotoxic activities (Badisa et al., 2003). In the current report, we investigated five species of the genus Salvia. against various cell lines. Salvia. is the largest genus of Labiatae family with more than 900 species existing throughout the world. Particular interest was paid to members of Salvia. genus due to the fact that many of these plants are aromatic and provide essential oils. One interesting species of Salvia. genus is S. miltiorrhiza. Bunge, which has unique pigments called tanshinone (unusual diterpenes) with a wide range of biological activities such as protection against myocardial ischemia (Yagi et al., 1994), antiplatelet aggregation (Onitsuka et al., 1983), antifungal (Honda et al., 1988), and antitumor activities (Ryu et al., 1997). Salvia. plants exhibit a wide range of biological activities, such as antioxidant (Cuvelier et al., 1994), antifungal (Daferera et al., 2000; Pitarokili et al., 20022003), antimicrobial (Newall et al., 1996; Tzakou et al., 2001), and antiviral activities (Okuno et al., 1994; Watanabe et al., 2000). Moreover, many studies have been reported on antitumor activities of Salvia. species, especially for S. miltiorrhiza., under in vitro. conditions (Ryu et al., 1997; Topcu et al., 1997; Sung et al., 1999; Liu et al., 2000; Chen et al., 20012002; Ulubelen, et al., 2003).

In the current study, in continuation of our interest in Greek plants with possible cytotoxic activity preliminary results on Salvia. species led us to the investigation of eight crude extracts of five Salvia. species growing wild. They were tested against four human cancer cell lines and a normal mouse cell line under in vitro. conditions.

Materials and Methods

Plant materials and preparation of extracts

The plant samples were collected during their flowering stage in 2001. The collection site of each plant is shown in Table 1. Voucher specimens are deposited in the Herbarium of the University of Athens (ATHU). The aerial parts of the plants were dried at room temperature and then reduced to coarse powder. Five grams of each sample were extracted with methanol for 1 day, and then the solvent was evaporated in vacuo. at 40°C. The obtained residues were reconstituted in 100% DMSO at 250 mg/ml, and stored under frozen for further use.

Table 1. Brine shrimp lethality test for extracts of Salvia. species.

Plant name	Sample no.	Collection site	LC50 (μg/ml)
Salvia fruticosa. L.	1	Kalymnos Island	205.7
Salvia fruticosa. L.	2	Rhodos Island	184.9
Salvia fruticosa. L.	3	Sifnos Island	> 1000
Salvia fruticosa. L.	4	Mt. Yuhta, Crete Island	171.8
Salvia pomifera. subsp. calycina. (Sm.) Hayek	5	Mt. Parnitha, Prefecture Attiki	291.7
Salvia ringens. Sibth. & Sm.	6	Mt. Parnitha, Prefecture Attiki	183.5
Salvia sclarea. L.	7	Mt. Pindos, Prefecture Trikala	128.8
Salvia verbenaca. L.	8	Zante Island	> 1000

The brine shrimp assay was done initially at 1 mg/ml for all eight samples in duplicate vials. The active samples that killed all shrimps at this dose were further tested at four different concentrations (0.1 to 0.4 mg/ml). After 24 h, the live shrimps were counted and the ED₅₀ was determined with 95% confidence intervals for each extracts by a Finney computer program.

Maintenance of cell lines

The following human cancer cell lines were purchased from the American Type Culture Collection (ATCC, Rockville, MD, USA) and used in the current investigation: human colon adenocarcinoma (HCA), human hepatoblastoma (HepG2), human breast cancer cells (MCF-7), and human pancreatic carcinoma (HPC). Apart from these cell lines, a normal mouse adipose areolar cell line (NCTC clone 929) was used as a control cell line for comparison.

HCA and HepG2 cells were cultured as described earlier (Badisa et al., 2000). In brief, these cultures were maintained in minimum essential medium (alpha modification) supplemented with 10% cosmic calf serum, gentamicin (0.5 µg/ml) and antibiotic-antimycotic solution, which contains penicillin (100 units/ml), streptomycin (100 µg/ml), and amphotericin B (0.25 µg/ml). MCF-7 cell line was maintained in Eagle's minimum essential medium supplemented with 10% fetal bovine serum, 0.01 mg/ml bovine insulin, penicillin (100 units/ml), and streptomycin (100 μg/ml). HPC cell line was maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum, penicillin (100 units/ml), and streptomycin (100 µg/ml). Normal mouse cell line was maintained in RPMI 1640 supplemented with 10% equine serum, penicillin (100 units/ml), and streptomycin (100 µg/ml). All cell lines were maintained as mono layer cultures in T-75 culture flasks in a humidified chamber at 37°C, 5% CO₂ in air in an incubator.

Brine shrimp lethality assay

The brine shrimp eggs (Artemia salina. Leach) were hatched in artificial seawater at room temperature and were used after 48 h. The lethality assay was performed as described by Meyer et al. (1982) and modified by McLaughlin (1991). In brief, the artificial seawater (5 ml) was added to each glass vial. Then, different volumes of test samples, which were present in 100% DMSO at 250 mg/ml, were added to the vials to achieve the required concentrations (1.0, 0.4, 0.3, 0.2, and 0.1 mg/ml) and mixed well. Ten shrimps were added to each vial. The shrimps in artificial seawater alone and with DMSO (maximum final concentration = 0.4%) in artificial seawater served as controls. After 24 h, the number of live shrimps per dose was recorded. The values of effective dose (ED₅₀) that killed 50% of shrimps were determined with 95% confidence intervals for each extract by a Finney computer program.

Treatment of cells with crude extracts

Treatment with crude plant extracts was done essentially as described earlier (Couladis et al., 2002). The preliminary studies with all crude extracts were done by continuous exposure of extracts at 0.1 mg/ml against all human cancer cell lines for 24 h. The extracts that killed > 75% cells at this concentration were further tested at five doses (10, 25, 50, 75, and 100 µg/ml) in order to determine the lethal concentration where 50% cells were killed (LC₅₀). For this purpose, HCA, HepG2, HPC, and normal mouse cells were plated at a starting cell density of 1 × 10⁴ cells per well of 96-well titer plates in a total volume of 196 µl. Then, the cells were allowed to adhere to the surface overnight in the incubator. Typically, by the following day, the cells were found to be of about 65–75% confluence. The crude extracts were diluted in the culture medium in order to achieve required working stocks and added in a total volume of 4 µl per culture well under sterile conditions. Each concentration was done in quadruplicate, and a total of five different concentrations (10 to 100 µg/ml) were tested. In case of MCF-7, the cells were plated in a 24-well culture plates at a density of 4 × 10⁴ per well in a total volume of 0.8 ml per well. The cells were allowed to attach to the surface overnight. On the next day, the samples that were in 100% DMSO were diluted in the culture medium under sterile conditions and added to the cultures to achieve the correct concentration. Two wells were used per concentration. Apart from these cell lines, a normal mouse cell line was used to study the selectivity of crude extracts for cytotoxicity at 0.1 mg/ml for 48 h in 96-cell titer plates.

In all cell lines, untreated and DMSO treated (0.4%) cells served as controls. DMSO at 0.4% did not alter the cell viability. All culture plates were incubated at 5% CO₂, 37°C for 48 h with the culture plates capped in an ordinary way. All studies were conducted at least two times independently under identical conditions.

Evaluation of cytotoxicity of crude extracts

After 48 h of treatment, the cytotoxicity of crude herbal extracts was evaluated by a dye uptake assay by using crystal violet as described previously (Couladis et al., 2002). In brief, after the end of incubation, 100 µl of 0.25% glutaraldehyde was added per well and incubated at room temperature for 30 min. The plates were washed in tap water and dried. Then the cells were stained with 100 µl of 0.1% crystal violet for 15 min. The excess dye was removed by washing the plates with tap water. Finally, 100 µl of 50 mM unbuffered sodium phosphate solution (monobasic) containing 50% ethyl alcohol was added to each well. The plates were gently vortexed and read on a microplate reader at 540 nm. From the average absorbance values of the treated and controls, the percent cells killed were determined by the following equation: [1 − (T./C.)] × 100, where T. is the average absorbance values of treated cells, and C. is the average absorbance values of control cells. A graph was plotted between the concentration of the extract and the percent cell death. The concentration of the crude extract that killed 50% cells (LC₅₀) was calculated.

Results and Discussion

In the current investigation, eight crude extracts of five Salvia. species were evaluated against brine shrimps and various cell lines for cytotoxic activities. According to Meyer et al. (1982), the crude extracts with ED₅₀ values less than 1000 µg/ml against brine shrimps are considered to be active. The tested extracts, except S. fruticosa.L. (Sifnos collection) and S. verbenaca.L. (Zante collection), presented ED₅₀ values ranging from 128.8 to 291.7 µg/ml (Table 1) and therefore were considered highly cytotoxic. Because all samples were dissolved in DMSO and then tested against shrimp, a corresponding DMSO control was taken (0.4% final concentration). In none of the control vials were the shrimp killed. In fact, it was noticed that DMSO up to 2.3% did not kill the shrimp. However, DMSO at 4.5% caused 60% shrimp death in 24 h (Badisa et al., 2003).

Based on the promising brine shrimp assay results with these samples, all Salvia. extracts were tested against HCA, HepG2, MCF-7, and HPC cancer cell lines. Apart from these cell lines, a normal mouse adipose areolar cell line (NCTC clone 929) was used as a control cell line for comparison. The highest dose of the samples tested in cancer cell lines (0.1 mg/ml) was used as a fixed concentration to test these extracts on normal mouse cell line. This may help to observe differential cytotoxic activity between cancer and normal cells (Badisa et al., 2003). The cytotoxic results on cell lines are shown in Table 2.

Table 2.. Cytotoxic activity of Salvia. extracts against human cancer cell lines in vitro..

Sampleno.	HCA LC50(μg/ml)	HepG2 LC50(μg/ml)	MCF-7 LC50(μg/ml)	HPC LC50(μg/ml)	Normal mousecell line,tested at 0.1 mg/ml(% dead)
1	60.4	68.9	43.1	42.2	93
2	> 75	> 75	41.1	> 75	80
3	> 75	> 75	> 75	64.1	76
4	40.1	> 75	42.3	50.3	89
5	> 75	> 75	> 75	> 75	25
6	> 75	> 75	> 75	> 75	21
7	> 75	> 75	> 75	> 75	4
8	> 75	> 75	> 75	> 75	12

The cell lines were treated continuously with five different concentrations (10, 25, 50, 75, and 100 µg/ml) of crude plant extracts for 48 h in the incubator. Cytotoxicity was evaluated (Couladis et al., 2002) and the percent dead cells were calculated in comparison to DMSO and untreated controls. DMSO control (maximum concentration = 0.4%) did not alter the cell viability. The concentration of the extract that killed 50% (LC₅₀) was determined from the graph, plotted between the concentration and the percent cell death. All values represent at least two independent experiments.

In the current study, four of the tested Salvia. species [S. pomifera. (Sm.) Hayek, S. ringens. Sibth. & Sm., S. sclarea.L., and S. verbenaca.L.] were found to have LC₅₀ values greater than 75 µg/ml in all cell lines tested and therefore considered to be inactive. As expected, these samples caused a marginal cell death, ranging from 4% to 25% in normal mouse cell line. It was observed that all S. fruticosa. (Greek sage) samples, collected from different locations in Greece, were cytotoxic to at least one or more of the human cancer cell lines tested. These samples were found to be cytotoxic to normal mouse cell line also, where more than 75% of the cells were killed at 0.1 mg/ml. This clearly indicates the lack of differential cytotoxic activity in cancer and normal mouse cell lines with these crude extracts. However, there was a significant differential cytotoxic activity among human cancer cell lines with these samples. For example, S. fruticosa. (Kalymnos collection) was found to be significantly more cytotoxic ( ∼ 30%) in MCF-7 (LC₅₀ 43.1 µg/ml) and HPC (LC₅₀ 42.2 µg/ml) cell lines than in HCA (LC₅₀ 60.4 µg/ml) and HepG2 (LC₅₀ 68.9 µg/ml) cell lines. Interestingly, the other two S. fruticosa. samples (Rhodos and Sifnos collection) showed specific cytotoxic activities in MCF-7 (LC₅₀ 41.1 µg/ml) and HPC (LC₅₀ 64.1 µg/ml) cell lines, respectively. Another S. fruticosa. sample (Crete collection) was found to be cytotoxic to HCA (LC₅₀ 40.1 µg/ml), MCF-7 (LC₅₀ 42.3 µg/ml), and HPC (LC₅₀ 50.3 µg/ml) cell lines, while it remained inactive in HepG2 cells (LC₅₀ > 75 µg/ml). It is not surprising that the same species, grown at different locations, vary in the cytotoxic activities, as it is known that different climatological and handling factors may result in variation in the chemical composition. It is reported that the essential oils content in members of Labiatae family is affected by light (Circella et al., 1995; Peer & Langenheim, 1998; Johnson et al., 1999), nutrition availability (Hornok, 1983; Franz, 1989), and seasons (Kokkini et al., 1997). As it is well-known, Salvia. plants are rich in volatile oils which possess antioxidant, antifungal, antimicrobial, antiviral activities as well as a cytotoxic activity (Foray et al., 1999). In the current study, the observed variation in cytotoxic activities of the crude extracts within a particular cancer cell line may be attributed to these factors. Another parameter worth considering for this variation is the sensitivity of the cell line studied. Comparison of results of all S. fruticosa. samples indicates that HepG2 cells are much more resistant than HCA cells, whereas MCF-7, HPC, and normal mouse cells are more sensitive.

Comparison of Tables 1 and 2 revealed a degree of correlation of cytotoxic results between brine shrimp and the cell lines tested. For example, 75% of Salvia. samples tested were active against brine shrimp, while 50% of them were active against the cell lines tested. This observation is in agreement with previous report about correlation of cytotoxic activities between brine shrimps and some human cancer cell lines (Anderson et al., 1991).

The Salvia. plants rich in volatile oils are considered to be a good source of natural antioxidants. The antioxidant activities for some of the studied species have been reported recently (Couladis et al., 2003). It is interesting to observe that some of the extracts in the current study, for example S. fruticosa., have both cytotoxic and antioxidant activities (Couladis et al., 2003). This may indicate that the mode of action for cytotoxicity in the tested cell lines by the active extracts is not, at least, through free-radical damage in these cells. In many cases in the literature, the same compound is responsible for both antioxidant and cytotoxic activity (Kubo et al., 1996; Ramsewak et al., 2000). As crude extracts were used in this study, it is not clear if the cytotoxic and antioxidant activities were due to the same or different factors.

In conclusion, all four different S. fruticosa. samples tested in the current study caused significant cytotoxic activities either on one or all four human cancer cell lines tested. It is also significant to note that all four S. fruticosa. samples caused a high percentage of death to normal mouse cells. This observation is surprising in view of the fact that Greek sage is used as a folk medicine in Mediterranean area (Wichtl, 1994). But it may be important to note that as in vitro. observations are entirely different, the extrapola of these results to the in vivo. situation should be done with caution. Further studies, however, are required to ascertain the in vivo. implications of these species and especially of the most active S. fruticosa. sample.

Acknowledgments

The authors are grateful to Professor A. Yannitsaros (Institute of Systematic Botany, Department of Biology, University of Athens) and Lecturer Th. Constantinidis (Institute of Systematic Botany, Department of Biotechnology, Agricultural University of Athens) for the identification of the specimens.