In Vitro Cytotoxic Activity against Breast, Cervical, and Ovarian Cancer Cells and Flavonoid Content of Plant Ingredients Used in a Selected Thai Traditional Cancer Remedy: Correlation and Hierarchical Cluster Analysis

This study aimed to investigate in vitro cytotoxic activity of selected plant ingredients from a traditional Thai remedy for the treatment of cancer patients against cancer cells occurring in women such as MCF-7 (breast cancer), SKOV3 (ovarian cancer), and HeLa (cervical cancer) cell lines. The plants and the remedy were macerated with 95% ethanol and boiled in water. Cytotoxic activity of the extracts was analyzed by SRB assay. Total flavonoid contents of the extracts were determined and their correlation with cytotoxic activity was evaluated. The hierarchical cluster analysis (HCA) was used to classify the extracts by their cytotoxic characteristics. A total of 66.7% of the plants was active against the tested cancer cell lines. Among the 44 plants in the remedy used for cancer treatment, nine plants that are also used in Thai cuisine exerted significant cytotoxicity against tested cancer cell lines. Eleven plants in the remedy were active against at least one of the tested cancer cell lines. All extracts were grouped into three groups and illustrated as heat map and hierarchical dendrogram. Total flavonoid content showed weak or no correlation with cytotoxic activity. A. dahurica, F. albopurpurea, and T. indica selectively exerted potent cytotoxic activity against MCF-7 with SI value more than 6. A. galanga, P. amarus, L. striatum, H. indicum, and F. vulgare exerted moderate cytotoxicity to all tested cell with low toxicity to normal cells. The correlation and HCA performed in this study provided an alternative way to investigate biological activities of plant ingredients in polyherbal traditional remedies.


Introduction
Cancers of breast, cervical, and ovarian are common cancers occurring in the female population. Breast cancer is the most common and the most leading cause of cancer death among women worldwide. Cervical and ovary cancers are the fourth and seventh most common, respectively [1]. Although each cancer has its own established treatment guidelines, the conventional treatments are surgery, radiation, and chemotherapy. e current treatments, particularly anticancer drugs, may be inadequate for patients due to some severe side effects arising from toxicity on normal cells or tissues. ese side effects of chemotherapy often affect patient's quality of life [2]. erefore, novel substances or plant extracts need to be developed for the treatment and prevention of cancers. Herbal medicines for cancer treatment are known to cancer patients. ese medicinal plants have become increasingly recognized either as alternative medicines for cancer treatment or dietary supplements for cancer prevention. In ai traditional medicine (TTM) prescribed by traditional doctors, several medicinal plants are combined as remedies [3,4]. TTM is a cultural heritage and indigenous wisdom that has been used by ai people for over a thousand years. It utilizes several herbs rather than a single herb to design a remedy for cancer treatment.
e selected TTM remedy in this study was a cancer remedy prescribed at the Jitmeatta Mercy Foundation for Cancer Patients of ailand (JFCT), Phetchaburi province, ailand. is remedy is composed of 44 plants mixed by equal weight. is traditional cancer remedy was prepared by the decoction method and it was used to treat cancer patients more than 30 years ago. Previous research showed that almost all cancer patients treated at JFCT were women, so this research pointed to prove that this cancer remedy can cure cancers in women [5].
Phytochemical compounds in herbal medicines play an important role in the expression of biological and pharmacological activities. Flavonoids, a group of phenolic compounds, provide potential benefits in health promotion and prevention of several chronic diseases including cancer. Due to the variation of substitutions on the skeletal structure of flavonoids, an extremely diverse range of flavonoid derivatives in herbal medicines have been discovered [6]. Many flavonoids showed cytotoxic activity and antitumor, such as epigallocatechin gallate and quercetin [7].
us, the aim of this study was to investigate in vitro cytotoxic activity of the extracts of this TTM remedy and its 44 plant ingredients in this remedy against commonly occurring cancers in women, i.e., MCF-7 (breast cancer), SKOV3 (ovarian cancer), and HeLa (cervical cancer). e total flavonoid content of the extracts was also studied, and its correlation with cytotoxic activity was also evaluated. e statistical cluster analysis was used to visualize the cytotoxic pattern of plants and remedies. is study also evaluated an alternative method for screening the cytotoxic activity of plant ingredients in polyherbal remedies used in traditional medicines. e correlation of cytotoxic activity with total flavonoid content may be used in the search for cytotoxic compounds from flavonoids against cancers occurring in women in the future.

Preparation of Plant and TTM Extracts.
Each plant was washed, thinly sliced, dried in an oven at 45°C, and powdered. Dried plants (1000 g) were macerated with 3.0 L of 95% ethanol for 3 days, filtered, and evaporated to dryness. e selected traditional ai cancer remedy was prepared according to the proportion used at the JFCT foundation by combining an equal weight of each dried plant, mixed together, and powdered. e resulting remedy was extracted by two methods, maceration (as described above), and decoction. e decoction method was carried out by boiling the remedy (1000 g) with an equal amount of water for 15 minutes and the aqueous extract filtered and evaporated to dryness with a freeze drier. e ethanolic extract of TTM was named TTE and the aqueous extract obtained from decoction was named TTW. e %yield was calculated (Table 1) and all extracts were kept at −20°C until use. MCF-7 and HeLa cell lines were cultured in minimum essential medium (MEM). SKOV-3 was cultured in Roswell Park Memorial Institute medium 1640 (RPMI-1640) and HaCat was cultured in Dulbecco's Modified Eagle's Medium (DMEM). All cultured media were supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin. e cells were incubated in 5% CO 2 at 37°C and 95% humidity.

In Vitro
Assay for Cytotoxic Activity. Cytotoxicity of the extracts was investigated by using sulphorhodamine B (SRB) assay according to the method previously described [8,9]. Briefly, cells were seeded in a 96-well microtiter plate and incubated for 24 hours. Each extract was diluted by DMSO to produce stock solution at a concentration of 10 mg/ml except TTW, which was diluted by water. e stock solution was diluted serially by culture medium to produce a working solution (2, 20, 100, and 200 μg/ml). e seeded cells were treated with various concentrations of working solution of plant and remedy extracts and incubated for 72 hours. e final concentrations of extract in the seeded cell were 1, 10, 50, and 100 μg/ml and the maximum final concentration of DMSO in the seeded cell was 1%v/v. e medium was replaced with cell culture media and the plates were continuously incubated at 37°C for three days in order to observe cytotoxicity after a recovery period. e cells were fixed with 40% trichloroacetic acid and the plates left at 4°C for 45 min. e media were removed and the cells were stained with 0.4% SRB in acetic acid. e %survival was determined by colorimetric measurement at 492 nm and calculated by the following equation. e half inhibitory concentrations (IC 50 ) of the extracts were calculated by using GraphPad Prism software (CA, USA)

Total Flavonoid Content.
Total flavonoid content was measured by the aluminum chloride colorimetric assay [10]. Briefly, an aliquot (1 mL) of the extracts or standard solution of quercetin (20, 40, 60, 80, 100, 120, and 240 mg/L) was added to a volumetric flask. To the flask were added 75 μL of 5% NaNO 2 and 150 μL of 10% AlCl 3 . After 5 min, 500 μL of 1 M NaOH was added, and the volume was adjusted by H 2 O. e solution was mixed well, and the absorbance was measured against the prepared reagent blank at 510 nm. e concentration of total flavonoid content in the test samples was calculated from the calibration plot (Y � 0.0162x + 0.0044, r 2 > 0.999) and expressed as mg quercetin equivalent/g of dried plant. All the determinations were carried out in triplicate.

Statistical Analysis.
All experiments were carried out by three independent experiments, and the data were reported as the mean and standard error of means. e correlation of total flavonoid content and cytotoxic activity was plotted and the correlation coefficient (r) calculated by using  15 extracts showed no cytotoxic activity with IC 50 values more than 100 μg/mL. e selective index (SI) of the extracts was calculated. SI is the ratio between IC 50 values of HaCaT (normal cell) and cancer cell lines. e greater the SI value, the more selective it is. SI value more than 2 indicates that the herbal extract is toxic against cancer cells but less toxic to normal cells. SI value less than 2 indicates the general toxicity of the extract [12]. Based on this criterion, the extracts of. A. dahurica, F. albopurpurea, M. siamensis, C. halicacalum, C. papaya, K. galanga, P. cablin, and T. indica were remarkably selective to breast cancer (MCF-7) cells. Plant extracts which exerted activities against all tested cancer cells were from A. galaga, L. striatum, P. amarus, and H. indicum.

Total Flavonoid Content and Correlation with Cytotoxic
Activity.
e total flavonoid content of the extracts is shown in Table 3. C. aromatica showed the highest content (259.7 ± 3.21 mg quercetin eq./g) followed by A. sinensis (259.70 ± 2.59 mg quercetin eq./g) and A. dahurica (227.95 ± 5.69 mg quercetin eq./g). e ethanolic extract of the ai traditional remedy showed total flavonoid content of 105.67 ± 4.33 mg quercetin eq./g. e correlation between flavonoid content and cytotoxic activity of the extracts was investigated. Total flavonoid contents of the extracts were plotted against their cytotoxic activities, and the correlation coefficients (r) were evaluated. e results showed that there was no correlation between the total flavonoid content of the extracts and their cytotoxic activity against all tested cell lines (r < 0.35) (Figure 1). However, in MCF-7, when some outlier data were excluded, there was a strong correlation between flavonoid content and cytotoxicity with r value −0.81 (Figure 2).

Hierarchical Cluster Analysis.
HCA is a multivariate statistical method to evaluate the pattern of data, which is classified into clusters according to their similar characteristics.
e similarity or dissimilarity of the extracts is illustrated, as shown in the dendrogram (Figure 3). In this study, 44 plant extracts and two ai traditional remedy extracts were classified by HemI software based on their cytotoxic activity against all tested cancer cells. e euclidean metric was selected to evaluate the similarity of the cytotoxic pattern. e heat map was generated to visualize the intensity of cytotoxic activity, and the linkages of similar extracts were shown in the horizontal dimension. As shown in Figure 3, HCA classified 46 extracts into three main groups. e extracts expressing low cytotoxicity against cancer cells were clustered into group I. Group II was composed of the extracts that selectively exhibited cytotoxicity against MCF-7. e extracts showing cytotoxicity against two or more cancer cells were clustered in group III.

Discussion
ere has been an increasing interest in herbal medicine as an alternative treatment and prevention of cancer. Traditional ai medicine is an alternative healthcare approach utilizing medicinal plants as a component in the holistic treatment of cancer patients [3]. In this study, we investigated the cytotoxic activity of selected ai medicinal plants and remedy against cancer cells commonly occurred in women: breast (MCF-7), ovarian (SKOV3), and cervical cancer (HeLa). In the cytotoxic screening of 44 plants and two TTM extracts, the results were analyzed by HCA and visualized them into a heat map (Figure 3). ree major groups of the extracts were categorized by their cytotoxicity characteristics. Group I is composed of plant extracts which exerted less or no cytotoxic activity. Group II was the extracts that selectively exerted cytotoxicity against MCF-7. e extracts with moderate to high activity against all cancer cells were clustered into group III. With regard to the extracts obtained from the selected cancer remedy, TTW was not active against all tested cells, while TTE exerted moderate cytotoxicity against MCF-7.
ese results indicated that the cytotoxic compounds were extracted from the remedy in a small amount. However, the results of cancer treatment by TTM may occur by the accumulation of the cytotoxic compounds in the body when it was taken for long-term multiple doses. In TTM, many plants in remedy are used as many functions or biological activities related to cancer such as antiinflammation, increase immunology, antimicrobial, antioxidant activities. Another reason, TTM extract may help with another mechanism on cancer treatment such as increased immunology or angiogenesis, when this remedy was absorbed in the patient's body. us, this suggestion should be clarify further studied including in vivo and clinical trials investigation.
According to the heat map and HCA shown in Figure 3 along with IC 50 and SI values presented in Tables 1 and 3, A. dahurica, F. albopurpurea, and T. indica exerted potent activity against MCF-7 with IC 50 less than 20 μg/ml and SI more than 6. M. siamensis exerted potent cytotoxicity against MCF-7 with SI more than 2. e results indicated that these extracts might have particularly beneficial effects for use in breast cancer treatment and prevention. A. galanga, P. amarus, and L. striatum exerted moderate cytotoxicity (IC 50 > 20 μg/ml) against all tested cancer cells with low toxicity to normal cells (SI > 2). H. indicum and F. vugare also selectively exerted mild to moderate cytotoxicity against the cancer cells.
Flavonoids, a group of phenolic compounds found in several medicinal plants, have a diverse range of derivatives depending on the substituted groups on their skeletal structure.
is variety of flavonoids make the medicinal plants more valuable for their role in health promotion and prevention of several chronic diseases including cancer. In this study, we investigated the total flavonoid contents of the plants and analyzed their correlation with cytotoxicity against cancer cells. As shown in Figure 1, the total flavonoid content of the extracts exhibited weak or no correlation to cytotoxicity against all tested cancer cells with coefficient (r) less than 0.35. However, in MCF-7, when the outlier data were excluded and the correlation recalculated, some extracts showed a strong correlation between total flavonoid Evidence-Based Complementary and Alternative Medicine content and cytotoxicity with the coefficient of −0.81 ( Figure 2). is indicated that flavonoids found in plants such as T. indica, B. rotunda, and A. dahurica might be related to their cytotoxicity against MCF-7. Several flavonoids from T. indica have been reported, including vitexin, isovitrexin, orientin, and iso-orientin [13].
A previous report showed that vitexin and orientin exhibited mild to moderate cytotoxic activity against MCF-7 cell line with IC 50 more than 50 μg/ml [14]. Another report showed that the rhizome of B. rotunda contained flavonoids and chalcone compounds such as alpinetin, pinocembrin, cardamonin, pinostrobin, and panduratin A [15]. A cyclohexenyl chalcone, panduratin A, in B. rotunda showed potent cytotoxic activity against MCF-7 with IC 50 of 3.75 μg/ ml [16]. A. dahurica showed high flavonoid content and exerted potent cytotoxic activity against MCF-7. e major constituents of A. dahurica were furocoumarins, imperatorin, phelloptorin, and isoimperatorin [17]. Imperatorin and isoimperatorin exhibited low to moderate cytotoxic activity against L1210, HL-60, K562, and B16F10 cell lines [18]. Components found in M. siamensis are coumarins such as mammea A, mammea B, mammea E, and other derivatives. Mammea A/AA exerted potent cytotoxic activity against breast cancer (MDA-MB-231) [19]. In P. amarus, the       Evidence-Based Complementary and Alternative Medicine related to cytotoxicity against the cancer cells. Major chemical constituents found in C. aromatica were volatile terpenes, sesquiterpenes, and curcuminoids. ese compounds may react with the chemicals used in the total flavonoid assay leading to the overestimation of the values [21]. C. verum also showed high content of flavonoids, the major components being cinnamaldehyde and methyl eugenol [22].
Among the plants that showed high total flavonoid content, C. aromatica extract which possessed the highest flavonoid content exerted mild cytotoxic activity while A. dahurica exerted potent cytotoxicity. is indicated that the flavonoids in C. aromatica differed from those in A. dahurica, leading to the differences in their cytotoxic activity. In phytochemical reviews, the major constituent in C. aromatica is terpenes, while in A. dahurica are furocoumarins.
ese indicated that these compounds might cause overestimation of total flavonoid content, which underscores the limitation of this study. Hence, the determination of the total flavonoid content of the plants alone may be insufficient to describe the correlation with their biological activities. erefore, more in-depth phytochemical studies should be pursued by more powerful analytical methods and instruments.

Conclusion
Among the 44 plants in a selected cancer remedy used for cancer treatments, 17 plants are food ingredients in ai cuisine and nine of them exerted significant cytotoxicity against the tested cancer cell lines. Eleven plants normally used in TTM were active against at least one of the cancer cell lines, while 66.7% of all extracts were active against all the tested cell lines. e remaining plant extracts may not have any cytotoxic activity but may be necessary adjuvants according to the TTM theory which considers the correction of the imbalance of body functions as an important aspect in designing a medicine. e correlation and HCA studies provided an alternative way to investigate the biological activities of plant ingredients in polyherbal traditional remedies. is method was designed to predict the correlation between their active components and biological activities. However, the results may not show a strong correlation between the active components in traditional remedies or mixed herbs and their cytotoxicity against cancer cells. is is because only some herbs in the remedy possess cytotoxic effects while the others are included to balance other physiological functions in the body to improve patient's well-being. erefore, in vivo studies are necessary to elucidate the true efficacy of herbal medicines. Hierarchical cluster analysis can help interpret the results of in vitro studies of herbals extracts by classifying them into groups that show different degrees of cytotoxicity in relation to the content of their bioactive components.
Data Availability e data supporting the conclusions of the study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.