Bioassay-Guided Isolation of Two Flavonoids from Derris scandens with Topoisomerase II Poison Activity

Abstract

Derris scandens (ROXB.) BENTH. (Fabaceae) is used as an alternative treatment for cancer in Thai traditional medicine. Investigation of the topoisomerase II (Top2) poison of compounds isolated from this plant may reveal new drug leads for the treatment of cancer. Bioassay-guided isolation was performed on an extract of D. scandens stems using a yeast cell-based assay. A yeast strain expressing the top2-1 temperature-sensitive mutant was used to assay Top2 activity. At the permissive temperature of 25°C, yeast cells were highly sensitive to Top2 poison agents. At the semi-permissive temperature of 30°C, where enzyme activity was present but greatly diminished, cells displayed only marginal sensitivity. The bioassay-guided fractionation of the extract led to the isolation of two known isoflavones: 5,7,4′-trihydroxy-6,8-diprenylisoflavone (1) and lupalbigenin (2). These two compounds also displayed cytotoxicity against three different cancer cell lines, KB, MCF-7 and NCI-H187. In conclusion, Top2 poison agents from D. scandens are reported for the first time, substantiating the use of D. scandens in Thai traditional medicine for cancer treatment.

DNA topoisomerase II (Top2) is a nuclear enzyme that alters DNA conformation through the concerted breaking and re-joining of both strands of the DNA backbone.¹⁾ Top2 plays important roles in a wide variety of DNA-mediated processes and proliferation, which is characterised by a high level of Top2 expression.^2,3) Top2 is also essential for the survival of eukaryotic cells⁴⁾ and has recently emerged as the principal intracellular target for chemical intervention in the development of anticancer agents. Many drugs target Top2 by trapping the enzyme in an intermediary reversible complex with DNA, termed the “cleavable complex,” which prevents the final re-joining step of the reaction and results in increased DNA strand cleavage.^5,6) These drugs are often referred to as Top2 poisons because their anticancer activity occurs via the trapping of Top2 rather the inhibition of Top2 activity. Top2 poison agents include amsacrine, doxorubicin and etoposide.^7–9) They have been developed and clinically applied but have side effects, such as dose-limiting toxicities and multidrug resistance (MDR), resulting in failed treatment after initial effective therapy.¹⁰⁾ Therefore, current research is focused on the development of novel Top2-targeted drugs to overcome the current limitations.

In the development of novel Top2 poison drugs, natural products from plants may be valuable sources and may provide suitable lead compounds for the production of semi-synthetic active agents. A yeast cell-based assay was selected because it allows yeast mutations in Top2 to be easily assessed in terms of the drug sensitivity conferred.¹¹⁾ A yeast model system for in vitro analysis was previously developed by Nitiss et al.¹⁾ When new Top2 poison agents were sought using a yeast model, Derris scandens, which is used as an alternative treatment for cancer in Thai traditional medicine, was chosen for further study.¹²⁾ Methanol extracts from the stem of D. scandens showed strong Top2 poison activity. Surprisingly, prior to this study, the Top2 poison activity of D. scandens had never been reported. Therefore, we pursued research to isolate the active compounds and identify the exact Top2-inhibition mechanism of D. scandens.

The present study aimed to determine the chemical components of the Top2 poison in D. scandens extract using the bioassay-guided fractionation technique. Cytotoxicity against human cancer cell lines was also determined to evaluate the Top2 poison potential of the isolated active compounds.

MATERIALS AND METHODS

General Experimental

NMR spectra were recorded on a Bruker Avance DPX-300 FT-NMR spectrometer. Mass spectra were recorded on a Micromass LCT mass spectrometer (electrospray ionization-time-of-flight (ESI-TOF)-MS). TLC was conducted on a silica gel 60 F254 (Merck, Germany) precoated plate. Silica gel 60 No.7734, 70–230 mesh, and No. 9385, 70–230 mesh (Merck) were used for vacuum-liquid chromatography and column chromatography, respectively. Etoposide was purchased from Sigma-Aldrich (Sigma-Aldrich, Vienna, Austria).

Plant Extracts and Chemicals

D. scandens stems were purchased from a Thai folk medicinal herb store in Bangkok, Thailand, and identified by Associate Professor Thatree Phadungcharoen at the Department of Pharmacognosy and Pharmaceutical Botany, Chulalongkorn University, Thailand. Voucher specimens (SS-SPT006) were deposited at the Herbarium of Natural Medicine, Chulalongkorn University, Thailand. Dried powdered stems of D. scandens (3 kg) were extracted with methanol (MeOH) (12 L), at room temperature to yield a methanol extract (277.8 g) after solvent evaporation.

Bioassay-Guided Fractionation

The methanol extract (100 g) was partitioned with ethyl acetate (EtOAc) and water (4×300 mL) to obtain an EtOAc extract (87.9 g) and a water extract (10.9 g), respectively. Both extracts were tested using a yeast cell-based assay. Then, the EtOAc extract was subjected to vacuum-liquid chromatography on silica gel (n-hexane–EtOAc, gradient from 100% n-hexane to 100% EtOAc) to yield 7 fractions (F1–F7). All fractions were tested for Top2 poison activity using a yeast-cell based assay. Fraction F4 was selected for further study because it showed the highest Top2 poison activity. The separation of fraction F4 (4 g) was performed using column chromatography over silica gel (n-hexane and EtOAc, gradient from 100% n-hexane to 100% EtOAc), and resulting compounds identify to be 5,7,4′-trihydroxy-6,8-diprenylisoflavone (1; 280 mg, Rf 0.85, silica gel n-hexane and EtOAc 6 : 4) and lupalbigenin (2; 48.5 mg, Rf 0.5, silica gel n-hexane and EtOAc 6 : 4).

Yeast Strains

The Saccharomyces cerevisiae strains were kindly provided by Dr. John Nitiss (St. Jude Children’s Research Hospital, U.S.A.). The parental yeast strain was S. cerevisiae JN394 ise2 ura3-52 leu2 trp1 ade2 his7 rad52::LEU2.¹³⁾ Two mutant strains derived from JN394 were also employed: JN394top1⁻, which has a chromosomal deletion of the topoisomerase I (TOP1) gene,¹³⁾ and JN394top2-1, in which the wild-type TOP2 gene is replaced with the temperature-sensitive top2-1 mutant allele.⁴⁾ These two strains are isogenic to JN394 in all other respects.¹⁴⁾

Yeast Spot Test

Yeast spot tests were conducted as previously described.¹²⁾ Briefly, yeast spot tests were conducted in yeast peptone dextrose adenine (YPDA) medium. The cultures were adjusted to an OD₅₉₀ of 0.5 and were serially diluted 10-fold; then, 5 µL aliquots were spotted onto the YPDA plates. For the primary screening of each plate, the fractions were diluted to 100 µg/mL. For the subsequent screening of bioactive compounds, these compounds were serially diluted from 200 µg/mL to 100, 50, 25,10, and 5 µg/mL. Etoposide was used as a positive control, and cells incubated with dimethyl sulfoxide (DMSO) alone were used as a vehicle control. The final concentration of DMSO never exceeded 2% (v/v) of the culture medium. Plates for determining cell viability were incubated at the optimal temperature for cell growth (25°C for temperature-sensitive TOP2 mutants; 30°C otherwise) for 3 to 4 d. The drug sensitivity was determined by comparing the survival of colonies in the extract culture with that in the drug-treated culture. Then, the colonies were photographed. All experiments were performed at least three times.

Cytotoxicity Assay

Cytotoxicity assays against human epidermoid carcinoma of the oral cavity (KB), breast adenocarcinoma (MCF-7) and human small cell lung carcinoma (NCI-H187) cell lines were performed with the Resazurin Microplate Assay (REMA).¹⁵⁾ Cells at a logarithmic growth phase were harvested and diluted to 7×10⁴ cells/mL for KB and 9×10⁴ cells/mL for MCF-7 and NCI-H187 in fresh medium. Then, 5 µL of test samples diluted in 5% DMSO and 45 µL of cell suspensions were added to 384-well plates and incubated at 37°C in a 5% CO₂ incubator. After the incubation period (3 d for KB and MCF-7 and 5 d for NCI-H187), 12.5 µL of 62.5 µg/mL resazurin solution was added to each well, and the plates were then incubated at 37°C for 4 h. The fluorescence signal was measured using a SpectraMax M5 multi-detection microplate reader (Molecular Devices, U.S.A.) at excitation and emission wavelengths of 530 and 590 nm, respectively. The cell growth inhibition percentage was calculated by using the following equation: % inhibition=[1−(FU_T/FU_C)]×100, where FU_T and FU_C represent the mean fluorescence units in treated and untreated conditions, respectively. Dose–response curves were plotted from 6 concentrations of 2-fold serially diluted test compounds and the sample concentrations that inhibit cell growth by 50% (IC₅₀) were determined using SOFTMax Pro software (Molecular Devices).

A cytotoxicity assay against Vero cells (African green monkey kidney) was performed with a green fluorescence protein (GFP)-based assay.¹⁶⁾ The assay was carried out by adding 45 µL of cell suspension at 3.3×10⁴ cells/mL to each well of the plates containing 5 µL of the test compounds previously diluted in 0.5% DMSO. After 4 d of humidified incubation at 37°C with 5% CO₂, fluorescence signals were measured using a SpectraMax M5 microplate reader (Molecular Devices) in the bottom reading mode with excitation and emission wavelengths of 485 and 535 nm, respectively. IC₅₀ values were calculated from dose–response curves, using 6 concentrations of 2-fold serially diluted samples with SOFTMax Pro software (Molecular Devices). Etoposide and 0.5% DMSO were used as positive and negative controls, respectively.

Statistical Analysis

Data are presented as the mean±standard deviation (S.D.) from at least three independent experiments. Sigmoidal dose responses were calculated to identify the IC₅₀ of the tested compounds.

RESULTS

Bioassay-Guided Fractionation of D. scandens Yielded Two Bioactive Compounds

The methanol extract of the stems of D. scandens was partitioned by its solubility in different solvents to obtain EtOAc and water extracts. Then, the EtOAc extract, which showed stronger inhibitory activity toward Top2 in yeast assays than the water extract, was repeatedly subjected to isolation. Bioassay-guided fractionation was performed on the EtOAc extract to isolate the bioactive compounds. During the fractionation process, the fractions were tested for Top2 poison activity by a yeast assay. The growing cultures of mutant yeasts at 25 and 30°C on YPDA agar containing a fixed concentration of 100 µg/mL of seven fractions (F1–F7) of the crude EtOAc extract showed three active fractions: F3, F4, and F5 (Fig. 1). Fraction F4 showed the highest Top2 poison activity. F4 fraction was further purified to give 5,7,4′-trihydroxy-6,8-diprenylisoflavone (1) (280 mg) and lupalbigenin (2) (48.5 g) (Fig. 2). The structures of the isolated compounds (1 and 2) were identified based on their spectroscopic properties and comparison with previously reported data.¹⁷⁾ The two compounds and the crude extract at various concentrations, 5, 10, 25, 50, 100, and 200 µg/mL, were then tested for Top2 poison activity (Fig. 3). The results revealed that they showed Top2 poison activity at different concentrations. At 25°C, the JN394top2-1 strain was not able to grow in the presence of compounds 1 and 2 at 200 and 10 µg/mL, respectively. When the concentrations were decreased, cell death was observed but to a lesser extent. The JN394top1⁻ strains were as sensitive to the Top2 poison agents as the wild-type (JN394) at both 25 and 30°C.

Fig. 1. The in Vitro Top2 Poison Screening of the Controls and Chromatographic Fractions, F1–F7, from D. scandens Stems by a Yeast Cell-Based Assay

Five microliters of 10-fold serial dilution (indicated by triangles) of JN394, JN394t2-1 and JN394top1⁻ were spotted on YPDA plates containing DMSO at 2% (v/v), 25 and 50 µg/mL of etoposide, and 100 µg/mL of EtOAc extract and F1–F7. The plates were separately incubated at 25°C (left) or 30°C (right) for 72 h, and the cell growth was observed and photographed.

Fig. 2. Chemical Structures of 5,7,4-Trihydroxy-6,8-diprenylisoflavone (1) and Lupalbigenin (2)

Fig. 3. The in Vitro Top2 Poison Screening of Crude EtOAc Extract, 5,7,4′-Trihydroxy-6,8-diprenylisoflavone (1) and Lupalbigenin (2) from D. scandens Stems at Various Concentrations by a Yeast Cell-Based Assay

Five microliters of ten-fold serial dilutions (indicated by triangles) of JN394, JN394t2-1 and JN394top1⁻ was spotted on plates that contained various concentrations of crude EtOAc extract, and compounds 1 and 2. The plates were incubated separately at 25°C (left) or 30°C (right) for 72 h, and the cell growth was observed and photographed.

Evaluation of the Cytotoxicity of Isolated Compounds 1 and 2

The cytotoxicities of an extract and the isolated compounds against three cancer cell lines (KB, MCF-7, NCI-H187) and a normal cell line (Vero cells) are shown (Table 1). Compounds 1 and 2 exhibited cytotoxicity against KB oral cavity cancer cells with IC₅₀ values of 7.34 and 10.02 µM, against MCF-7 breast cancer cells with IC₅₀ values of 8.15 and 4.15 µM and against NCI-H187 small lung cancer cells with IC₅₀ values of 6.59 and 7.91 µM, respectively.

Table 1. Cytotoxicity of Isolated Compounds 1 and 2 against Cancer Cell Lines

Compound	IC50* (µM)
	KB	MCF-7	NCI-H187	Vero cells
1	7.34±0.03	8.15±1.21	6.59±0.07	15.84±0.06
2	10.02±0.12	4.15±0.57	7.91±1.12	9.66±1.25
Etoposide	0.71±0.07	8.11±0.29	0.07±0.38	15.36±0.21

Etoposide was used as a positive control. * Data represent the mean±S.D. from three different experiments.

DISCUSSION

D. scandens stems, which are used in Thai folk remedies that are alternative treatments for cancer, were further investigated regarding their Top2 poison activity using a yeast cell-based assay. A yeast-model system for the in vitro analysis of the interaction between Top2 and various anticancer inhibitors was previously developed by Nitiss et al.¹⁾ This system has been extensively used to analyse the effects of specific anticancer agents on Top2.^1,4,14,18) The colony survival of the JN394top2-1 strain, which carries the top2-1 allele encoding a temperature-sensitive topoisomerase, was compared after exposure to the plant extracts at 25 and 30°C; this comparison made it possible to investigate the compounds’ mechanism of action. Yeast strains carrying top2-1 were found to have an activity approximating that of the wild-type at 25°C but substantially reduced activity at 30°C. Consequently, the top2-1 yeast strains that proliferate similarly to wild-type (JN394) cells at the permissive temperature (25°C) are non-viable at 37°C and have sufficient Top2 activity to be viable at the semi-permissive temperature (30°C).¹⁾ To show that the yeast cell death was caused by Top2 poison and was not related to the activity of Top1, JN394top1⁻ strains were tested.¹³⁾ The strains were constructed by taking advantage of the fact that the TOP1 gene is not essential in yeast. JN394top1⁻ strains that completely lack Top1 express only Top2 confirm this result on the cellular level. JN394top1⁻ strains have the same sensitivity to Top2 poison agents as that of wild-type strains.

The assay showed that the extract of D. scandens had Top2 poison activity against the mutant yeast cells (Fig. 1). Previous reports have demonstrated the cytotoxicity of the crude extract of the plant against cancer cell lines.¹⁹⁾ However, no previous studies have examined the mechanism of action in detail. Therefore, D. scandens was selected for further study through Top2 activity-guided fractionation (Fig. 1) using several chromatographic techniques. The NMR and MS data were compared with the literature,¹⁹⁾ and the isolated compounds were identified as 5,7,4′-trihydroxy-6,8-diprenylisoflavone (1) and lupalbigenin (2) (Fig. 2). This is the first report on the Top2 poison activity of these two flavonoids isolated from D. scandens.

In a previous study, some flavones exhibited Top2 inhibition. For example, genistein has been shown to stabilise the covalent Top2-DNA cleavage complex and, thus, to function as a Top2 poison.²⁰⁾ Although the core structures of 1 and 2 are identical to that of genistein, compounds 1 and 2 have two prenyl groups in their structures. A previous study investigated molecular docking between prenylated flavonoids and Top2. The presence of the prenyl group is also important for Top2 inhibition. It was found that the Top2 inhibition activity of prenylated flavonoids is due to the prenyl group that binds to a hydrophobic pocket of Top2.²¹⁾ Taken together, the results indicate that the structures of compounds 1 and 2 contribute to the Top2 inhibitory activity. Moreover, this experiment also validated the mutant yeast system as an effective and useful model for identifying the cell-killing activity of Top2 poisons.

To investigate whether these compounds (1 and 2) mediated the DNA Top2 poison activity to induce cell death, the compounds were tested on human cancer cell lines (Table 1). Compared with compound 2, compound 1 showed higher cytotoxicity against KB and NCI-H187 cell lines but lower cytotoxicity against MCF-7. Both 1 and 2 showed cytotoxicity against Vero cells (IC₅₀=15.84 and 9.66 µM, respectively); however, compared with etoposide (IC₅₀=8.32 µM), a Top2-targeted drug, compounds 1 and 2 exhibited lower cytotoxicity. In a previous study, D. scandens extract was also shown to be cytotoxic against lung and prostate cancer cell lines.²²⁾ Another study reported that the ethanol extract of D. scandens showed a potent antimetastatic effect, particularly for squamous cell carcinoma (KKU-M139) and adenosquamous carcinoma (KKU-M213).¹⁸⁾

Top2 targeting agents can be separated into two distinct groups, poisons and catalytic inhibitors. Those that stimulate the formation of enzyme–DNA covalent complexes, are termed Top2 poisons, and those that interfere with this covalent intermediate are called catalytic inhibitors. In contrast to the Top2 poisons, the catalytic inhibitors act through a variety of mechanisms including DNA binding, ATP binding or hydrolysis, magnesium binding, and non-covalent linkages between enzyme and DNA.²³⁾ Although compounds 1 and 2 act by stabilising the cleavable complex to show anti-cancer activity, it is possible that compounds 1 and 2 could act as mediators of Top2 catalytic inhibitors. If so, detailed studies evaluating the step(s) of the Top2 catalytic cycle must be performed.

Natural product-derived compounds continue to provide a valuable and rich pool for the discovery of novel anticancer agents. The use of a bioassay-guided fractionation method can validate the traditional use of medicinal plant species and support the cultural and medicinal significance of these taxa.

Herein, for the first time, a bioassay-guided fractionation of D. scandens stems using a yeast model resulted in the isolation of two anti-cancer agent isoflavones: 5,7,4′-trihydroxy-6,8-diprenylisoflavone (1) and lupalbigenin (2). These results demonstrate the potential of compounds 1 and 2 as novel Top2 poison agents.

Acknowledgments

This work was supported by a Research Grant from the Ratchadaphiseksomphot Endowment Fund (CU-57-003-HR) of Chulalongkorn University. We also thank the Chulalongkorn University Centenary Academic Development Project for access to their facilities.

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1) Nitiss JL, Liu YX, Hsiung Y. A temperature sensitivity topoisomerase II allele confers temperature dependent drug resistance on amsacrine and etoposide: A genetic system for determining the target of topoisomerase II inhibitors. Cancer Res., 53, 89–93 (1993).
• 2) Robert J, Larsen AK. Drug resistance to topoisomerase II inhibitors. Biochimie, 80, 247–254 (1998).
• 3) McClendon AK, Osheroff N. DNA topoisomerase II genotoxicity and cancer. Mutat. Res., 623, 83–97 (2007).
• 4) Elsea SH, Osheroff N, Nitiss JL. Cytotoxicity of quinolones toward eukaryotic cells. Identification of topoisomerase II as the primary cellular target for the quinolone CP-115,953 in yeast. J. Biol. Chem., 267, 13150–13153 (1992).
• 5) Liu LF. DNA topoisomerase poisons as anti-tumor drugs. Annu. Rev. Biochem., 58, 351–375 (1989).
• 6) Nitiss JL. Targeting DNA topoisomerase II in cancer chemotherapy. Nat. Rev. Cancer, 9, 338–350 (2009).
• 7) Meczes EL, Marsh KL, Fisher LM, Rogers MP, Austin CA. Complementation of temperature-sensitive topoisomerase II mutations in Saccharomyces cerevisiae by a human TOP2 beta construct allows the study of topoisomerase II beta inhibitors in yeast. Cancer Chemother. Pharmacol., 39, 367–375 (1997).
• 8) Patel S, Sprung AU, Keller BA, Heaton VJ, Fisher LM. Identification of yeast DNA topoisomerase II mutants resistant to the antitumor drug doxorubicin: Implication for the mechanism of doxorubicin action and cytotoxicity. Mol. Pharmacol., 52, 658–666 (1997).
• 9) Hammonds TR, Maxwell A, Jenkins JR. Use of a rapid throughput in vivo screen to investigate inhibitors of eukaryotic topoisomerase II enzymes. Antimicrob. Agents Chemother., 42, 889–894 (1998).
• 10) Hu CX, Zuo ZL, Xiong B, Ma JG, Geng MY, Lin LP, Jiang HL, Ding J. Salvicine functions as novel topoisomerase II poison by binding to ATP pocket. Mol. Pharmacol., 70, 1593–1601 (2006).
• 11) Reid RJD, Benedetti P, Bjornsti MA. Yeast as a model organism for studying the actions of DNA topoisomerase-targeted drugs. Biochim. Biophys. Acta, 1400, 289–300 (1998).
• 12) Sangmalee S, Laorpaksa A, Sukrong S. A topoisomerase II poison screen of ethnomedicinal Thai plants using a yeast cell-based assay. J. Ethnopharmacol., 142, 432–437 (2012).
• 13) Nitiss J, Wang JC. DNA topoisomerase-targeting antitumor drugs can be studied in yeast. Proc. Natl. Acad. Sci. U.S.A., 85, 7501–7505 (1988).
• 14) Meng LH, He XX, Zhang JS, Ding J. DNA topoisomerase II as primary cellular target for salvicine in Saccharomyces cerevisiae. Acta Pharmacol. Sin., 22, 741–746 (2001).
• 15) O’Brien J, Wilson I, Orton T, Pognan F. Investigation of Alamar Blue (resazurin) fluorescent dye for the assessment of mammalian cell cytotoxicity. Eur. J. Biochem., 267, 5421–5426 (2000).
• 16) Hunt L, Jordan M, De Jesus M, Wurm FM. GFP-expressing mammalian cells for fast, sensitive, noninvasive cell growth assessment in a kinetic mode. Biotechnol. Bioeng., 65, 201–205 (1999).
• 17) Sekine T, Inagaki M, Ikegami F, Fujii Y, Ruangrungsi N. Six diprenylisoflavones, derrisisoflavone A–F, from Derris scandens. Phytochemistry, 52, 87–94 (1999).
• 18) Froelich-Ammon SJ, Patchan MW, Osheroff N, Thompson B. Topoisomerase II binds to ellipticine in the absence or presence of DNA. Characterization of enzyme–drug interactions by fluorescence spectroscopy. J. Biol. Chem., 270, 14998–15004 (1995).
• 19) Laupattarakasem P, Sripa B, Laupattarakasem W. Antimigration of cancer cells by Derris scandens on cholangiocarcinoma cells. Srinagarind. Med. J., 22, 339–345 (2007).
• 20) Schmidt F, Knobbe CB, Frank B, Wolburg H, Weller M. The topoisomerase II inhibitor, genistein, induces G2/M arrest and apoptosis in human malignant glioma cell lines. Oncol. Rep., 19, 1061–1066 (2008).
• 21) Cassidy CE, Setzer WN. Cancer-relevant biochemical targets of cytotoxic Lonchocarpus flavonoids: A molecular docking analysis. J. Mol. Model., 16, 311–326 (2010).
• 22) Itharat A, Ooraikul B. Research on Thai medicinal plants for cancer treatment. (Acharya SN, Thomas JE eds.) Research Signpost, India, pp. 287–317 (2007).
• 23) Wang JC. DNA topoisomerases as targets of therapeutics: an overview. Adv. Pharmacol., 29A, 1–19 (1994).