Antiproliferative Activity of (-)-Rabdosiin Isolated from Ocimum sanctum L.

Background:Ocimum sanctum L. (holy basil; Tulsi in Hindi) is an important medicinal plant, traditionally used in India. Methods: The phytochemical study of the nonpolar (dichloromethane 100%) and polar (methanol:water; 7:3) extracts yielded fourteen compounds. Compounds 6, 7, 9, 11, 12, and 13, along with the methanol:water extract were evaluated for their cytotoxicity against the human cancer cell lines MCF-7, SKBR3, and HCT-116, and normal peripheral blood mononuclear cells (PBMCs). Results: Five terpenoids, namely, ursolic acid (1), oleanolic acid (2), betulinic acid (3), stigmasterol (4), and β-caryophyllene oxide (5); two lignans, i.e., (-)-rabdosiin (6) and shimobashiric acid C (7); three flavonoids, luteolin (8), its 7-O-β-D-glucuronide (9), apigenin 7-O-β-D-glucuronide (10); and four phenolics, (E)-p-coumaroyl 4-O-β-D-glucoside (11), 3-(3,4-dihydroxyphenyl) lactic acid (12), protocatechuic acid (13), and vanillic acid (14) were isolated. Compound 6 was the most cytotoxic against the human cancer lines assessed and showed very low cytotoxicity against PBMCs. Conclusions: Based on these results, the structure of compound 6 shows some promise as a selective anticancer drug scaffold.

Despite its wide therapeutic range, special care should be taken in case of the use of Tulsi in conjunction with other prescribed medicines since it exhibits various drug interactions. For example, its concomitant use with anticoagulants, such as heparin, warfarin, aspirin, clopidogrel, etc., is contraindicated due to allergic reactions that may occur. In addition, Tulsi increases the activity of phenobarbital and consequently may stimulate uterine contractions; thus, its use during pregnancy and lactation is not recommended [18,19].

Plant Material
Aerial parts of O. sanctum L. were collected in flowering stage at Suriname, as previously described [21]. A voucher specimen (ATHS 093) has been deposited in the Herbarium of the Laboratory of Pharmacognosy, National and Kapodistrian University of Athens.

General Experimental Procedures
1 H, 13 C, and 2D NMR spectra were recorded in CDCl 3 20 D values were obtained in CHCl 3 or MeOH on a Perkin-Elmer 341 Polarimeter. FT-IR spectra were recorded on a Perkin Elmer PARAGON 500 spectrophotometer. UV spectra were recorded on a Shimadzu UV-160 A spectrophotometer according to Mabry et al. (1970) [26]. GC-MS analyses were performed on a Hewlett-Packard 5973-6890 system operating in EI mode (70 eV) equipped with a split/splitless injector (220 • C), a split ratio 1/10, using a fused silica HP-5 MS capillary column (30 m x 0.25 mm (i.d.), film thickness: 0.25 µm) with a temperature program for HP-5 MS column from 60 • C (5 min) to 280 • C at a rate of 4 • C/min and helium as a carrier gas at a flow rate of 1.0 mL/min. Vacuum liquid chromatography (VLC): silica gel 60H (Merck, Art. 7736) [27].

Extraction and Isolation
The initial extraction was previously described [21]. In brief, the aerial parts of O. sanctum L. (0.40 kg) were air-dried and finely ground, and then extracted at room temperature using dichloromethane and methanol, successively.
Part of the dichloromethane residue (11.9 g) was re-extracted at room temperature with ethyl acetate (EtOAc) and n-BuOH, yielding two fractions (A and B). Fraction A (7.8 g) was fractionated by VLC on silica gel using mixtures of cyclohexane and EtOAc of increasing polarity (  69.4 mg) was subjected to CC on silica gel as previously described to give 75 fractions; fraction 8 (1.3 mg) was identified as compound 10. Another part of the methanol extract (7.7 g) was redissolved in water and extracted at room temperature with EtOAc and n-BuOH, affording three fractions (MA-MC). MB (eluted with n-BuOH; 5.3 g) was subjected to RP 18 -MPLC using a H 2 O:MeOH gradient system (100% H 2 O→100% MeOH; steps of 10% MeOH) and yielded 11 fractions (MB 1 -MB 10 ). Fraction MB 3 (eluted with H 2 O:MeOH 80:20) was identified as compound 8 (13.6 mg).
It is notable that during the fractionation and isolation procedures, all extracts and subfractions were continuously monitored by analytical TLC and 1 H-NMR. All obtained fractions were concentrated to dryness under vacuum (30 • C) and placed in activated desiccators with P 2 O 5 until their weights were stabilized.
Compounds were prepared at a stock solution of 10.0 mg/mL in DMSO and the extract at 20.0 mg/mL in DMSO. Prior to their use, they were diluted in plain RPMI-1640. Cytotoxicity was evaluated by the MTT reduction assay [29], which determines the effect of treatment with an exogenously added agent on the viability of the cell population. Briefly, cells were plated in 96-well plates (Greiner Bio-One GmbH, Frickenhausen, Germany; 5 × 10 3 cells/well) and incubated at 5% CO 2 and 95% air at 37 • C for 24 h, in order to adhere. Further, cells were incubated with the compounds for 72 h at 37 • C in a 5% CO 2 incubator. The MTT reagent (Sigma-Aldrich, Darmstadt, Germany; 1 mg/mL in phosphate buffered saline (PBS); 100 µL/well) was added during the last 4 h of incubation. The formazan crystals formed were dissolved by adding 0.1 M HCl in 2-propanol (100 µL/well) and absorption was measured using an ELISA reader (Denley WeScan, Finland) at 545 nm with reference filter set at 690 nm. All cultures were set in triplicate, whereas cells incubated in complete medium or in medium containing the equivalent amount of DMSO, as well as cells incubated in the presence of doxorubicin (Sigma-Aldrich) were used as negative and positive controls, respectively. The half maximal inhibitory concentration (IC 50 ) was calculated according to the formula: 100(A 0 − A)/A 0 = 50, where A and A 0 are optical densities of wells exposed to the compounds and control wells, respectively.
The compounds were tested at a concentration range of 200.0 to 6.25 µg/mL and the extract at 750.0 to 1.25 µg/mL. Doxorubicin was used as a standard cytotoxic agent and showed IC 50 values ≤ 0.2 µM in all cell lines tested. All experiments were performed at least three times.

Flow Cytometry Analysis
MCF-7, SKBR3 and HCT-116 cells were incubated with compound 6 and analyzed with flow cytometry following staining with annexin V and propidium iodide (PI). Cells were plated into 24-well plates (Greiner Bio-One; 3 × 10 5 /mL; 2 mL/well), let adhere overnight, and incubated with the mean IC 50 value (80 µg/mL) and 40 µg/mL of compound 6 for 72 h. Cells were detached with 2 mM EDTA in Dulbecco's PBS (DPBS), harvested, centrifuged in cold PBS (1500 rpm; 5 min), and stained with the Annexin V-FITC Apoptosis Detection Kit (BioLegend, Fell, Germany; cat# 640914), according to the manufacturers' instructions. In brief, cells were resuspended in binding buffer, then annexin V-FITC (5 µL) and PI (10 µL; 0.03 µg/sample) were added, mixed, and incubated with the cells for 15 min in the dark at room temperature. The volume was adjusted to 500 µL with binding buffer and the cell suspension was immediately analyzed in a FACSCanto II (BD Biosciences, San Diego, CA, USA) using FACSDiva software (V7, BD Biosciences).

Cytotoxic Effect against Human Peripheral Blood Mononuclear Cells
Compound 6 was additionally assessed for its cytotoxicity against human peripheral blood mononuclear cells (PBMCs) isolated from healthy blood donors' peripheral blood as previously described [30]. Prior to blood draw, individuals gave their informed consent according to the regulations approved by the 2nd Peripheral Blood Transfusion Unit and Hemophiliac Centre, "Laikon" General Hospital Institutional Review Board, Athens, Greece. PBMCs were seeded in 24-well plates (5 × 10 5 /mL; 2 mL/well) and exposed to 2 concentrations of compound 6: 80 µg/mL and 40 µg/mL. PBMCs were collected, stained as described in 2.5 and analyzed by flow cytometry.
According to the literature, the taxonomic description of the genus Ocimum L. is still debatable. It is composed of three subgenera, namely subgenus Ocimum (comprising three sections: Ocimum, Gratissima and Hiantia), subgenus Nautochilus, and subgenus Gymnocimum. The species (O. sanctum L.) under investigation has been located in the subgenus Gymnocimum. This subgenus can be distinguished because of the existence of flavonoid glucuronides, which are found in plants of the subgenera Nautochilus and Ocimum [38]. Consequently, our work is in agreement with previous studies regarding the chemical profile of the subgenus Gymnocimum. Moreover, it was previously shown that 3-(3,4-dihydroxyphenyl) lactic acid is a precursor of the nonenzymatic synthesis of (S)-(-)-rosmarinic acid and (+)-rabdosiin [47], therefore its identification (compound 12) could be related to the biosynthesis of (-)-rabdosiin (6) [48].
Compound (-)-rabdosiin (6) (Figure 1) is a caffeic acid tetramer connected to a lignan skeleton. Originally, it has been isolated and identified from the stem of Rabdosia japonica, Labiatae [35], while both enantiomers (-)-rabdosiin and (+)-rabdosiin were later isolated from Macrotomia euchroma, Boraginaceae [49] and also from other plants of this family such as Lithospermum erythrorhizon [50] and Eritrichium sericeum [36]. Based on the fact that the entire fractionation and isolation procedures were continuously monitored by 1 H-NMR, the active compound 6 was not detected in other fractions (NMR data of 6 are provided as Supplementary Materials, Tables S1 and S2, Figures S1-S6). Consequently, being a minor compound of the plant, its activity could derive in synergy with other constituents.
Medicines 2019, 6, x FOR PEER REVIEW 5 of 10 sanctum L.) under investigation has been located in the subgenus Gymnocimum. This subgenus can be distinguished because of the existence of flavonoid glucuronides, which are found in plants of the subgenera Nautochilus and Ocimum [38]. Consequently, our work is in agreement with previous studies regarding the chemical profile of the subgenus Gymnocimum. Moreover, it was previously shown that 3-(3,4-dihydroxyphenyl) lactic acid is a precursor of the nonenzymatic synthesis of (S)-(-)-rosmarinic acid and (+)-rabdosiin [47], therefore its identification (compound 12) could be related to the biosynthesis of (-)-rabdosiin (6) [48]. Compound (-)-rabdosiin (6) (Figure 1) is a caffeic acid tetramer connected to a lignan skeleton. Originally, it has been isolated and identified from the stem of Rabdosia japonica, Labiatae [35], while both enantiomers (-)-rabdosiin and (+)-rabdosiin were later isolated from Macrotomia euchroma, Boraginaceae [49] and also from other plants of this family such as Lithospermum erythrorhizon [50] and Eritrichium sericeum [36]. Based on the fact that the entire fractionation and isolation procedures were continuously monitored by 1 H-NMR, the active compound 6 was not detected in other fractions (NMR data of 6 are provided as Supplementary Materials, Tables S1 and S2, Figures S1-S6). Consequently, being a minor compound of the plant, its activity could derive in synergy with other constituents. According to published data, rabdosiin and the similar caffeic acid derivatives have been suggested as potential anti-HIV and antiallergic agents. Moreover, studies showed that rabdosiin is an antioxidant factor (acting as an effective scavenger of reactive oxygen species), as well as a possible inhibitor of hyaluronidase and β-hexosaminidase release [51,52]. Nevertheless, to the best of our knowledge, the antiproliferative activity of rabdosiin is reported for the first time.

Antiproliferative Activityod of Secondary Metabolites of O. sanctum
Using the MTT dye reduction assay, the methanol:water extract (7:3) and 6 purified secondary metabolites (compounds 6, 7, 9, 11, 12, and 13) were screened for their cytotoxic/cytostatic activity against human breast and colon cell lines. Our results showed that the extract was cytotoxic against all cell lines, with an IC50 range of 45  2.12 to 57  14.14 g/mL (Table 1). Based on these data, we further proceeded to the screening of the isolated natural products 6, 7, 9, 11, 12, and 13 against MCF-7 cells which was the mostly affected cell line exposed to the methanol extract of Ο. sanctum L. The IC50 values calculated are presented in Table 1. Among the purified compounds, the most prominent was 6, which was further tested against SKBR3 and HCT-116 cells. Overall, compound 6 demonstrated a considerable cytotoxic activity, with IC50 values 75  2.12, 83 ± 3.54 and 84 ± 7.78 μg/mL against MCF-7, SKBR3, and HCT-116, respectively.  According to published data, rabdosiin and the similar caffeic acid derivatives have been suggested as potential anti-HIV and antiallergic agents. Moreover, studies showed that rabdosiin is an antioxidant factor (acting as an effective scavenger of reactive oxygen species), as well as a possible inhibitor of hyaluronidase and β-hexosaminidase release [51,52]. Nevertheless, to the best of our knowledge, the antiproliferative activity of rabdosiin is reported for the first time.

Antiproliferative Activityod of Secondary Metabolites of O. sanctum
Using the MTT dye reduction assay, the methanol:water extract (7:3) and 6 purified secondary metabolites (compounds 6, 7, 9, 11, 12, and 13) were screened for their cytotoxic/cytostatic activity against human breast and colon cell lines. Our results showed that the extract was cytotoxic against all cell lines, with an IC 50 range of 45 ± 2.12 to 57 ± 14.14 µg/mL (Table 1). Based on these data, we further proceeded to the screening of the isolated natural products 6, 7, 9, 11, 12, and 13 against MCF-7 cells which was the mostly affected cell line exposed to the methanol extract of O. sanctum L. The IC 50 values calculated are presented in Table 1. Among the purified compounds, the most prominent was 6, which was further tested against SKBR3 and HCT-116 cells. Overall, compound 6 demonstrated a considerable cytotoxic activity, with IC 50 values 75 ± 2.12, 83 ± 3.54 and 84 ± 7.78 µg/mL against MCF-7, SKBR3, and HCT-116, respectively. To analyze the type of cell death (apoptosis or necrosis) induced by compound 6 on MCF-7, SKBR3, and HCT-116 cells, cells were stained with annexin V which binds phosphatidylserine exposed on the surface of apoptotic cells and PI which intracellulary stains the DNA of necrotic cells. As shown in Figure 2, 80 µg/mL of compound 6 drove ca. 50% of all cells to apoptosis. Specifically, 44.9% of MCF-7 were annexin V+ and 12.3% annexin V+/PI+, suggesting that cells exposed to compound 6 underwent early apoptosis and a small percentage thereof late apoptosis/necrosis. Analogous percentages were obtained for SKBR3 (40.1% early apoptotic; 9.1% late apoptotic/necrotic) and HCT-116 (43.1% early apoptotic; 10.2% late apoptotic/necrotic) cells. When the same cell lines were exposed to 40 µg/mL of compound 6, the percentages of early apoptotic and late apoptotic/necrotic cells were reduced ca. by 50% (13.5-20.1% and 3.9-6.5%, respectively), suggesting that induction of apoptosis by compound 6 is concentration-dependent. To analyze the type of cell death (apoptosis or necrosis) induced by compound 6 on MCF-7, SKBR3, and HCT-116 cells, cells were stained with annexin V which binds phosphatidylserine exposed on the surface of apoptotic cells and PI which intracellulary stains the DNA of necrotic cells. As shown in Figure 2, 80 μg/mL of compound 6 drove ca. 50% of all cells to apoptosis. Specifically, 44.9% of MCF-7 were annexin V+ and 12.3% annexin V+/PI+, suggesting that cells exposed to compound 6 underwent early apoptosis and a small percentage thereof late apoptosis/necrosis. Analogous percentages were obtained for SKBR3 (40.1% early apoptotic; 9.1% late apoptotic/necrotic) and HCT-116 (43.1% early apoptotic; 10.2% late apoptotic/necrotic) cells. When the same cell lines were exposed to 40 μg/mL of compound 6, the percentages of early apoptotic and late apoptotic/necrotic cells were reduced ca. by 50% (13.5-20.1% and 3.9-6.5%, respectively), suggesting that induction of apoptosis by compound 6 is concentration-dependent. Based on the significant cytotoxic activity of compound 6 against cancer cell lines we further tested whether it may also be toxic against normal cells, i.e., PBMCs isolated from two different healthy blood donors. PBMCs were incubated for 24 h with the IC50 and the 1/2 concentration of 6, stained and analyzed by flow cytometry. Interestingly, the IC50 of compound 6 (80 μg/mL) induced Based on the significant cytotoxic activity of compound 6 against cancer cell lines we further tested whether it may also be toxic against normal cells, i.e., PBMCs isolated from two different healthy blood donors. PBMCs were incubated for 24 h with the IC 50 and the 1/2 concentration of 6, stained and analyzed by flow cytometry. Interestingly, the IC 50 of compound 6 (80 µg/mL) induced early and late apoptosis/necrosis in a small percentage of PBMCs (2.8% and 3.0% for donor 1; 4.3% and 3.1% for donor 2, respectively). At half concentration, the percentages were highly reduced and much less early apoptotic and late apoptotic/necrotic cells were detected (1.8% and 1.7% for donor 1; 2.1% and 1.9% for donor 2, respectively) ( Figure 3).
Medicines 2019, 6, x FOR PEER REVIEW 7 of 10 early and late apoptosis/necrosis in a small percentage of PBMCs (2.8% and 3.0% for donor 1; 4.3% and 3.1% for donor 2, respectively). At half concentration, the percentages were highly reduced and much less early apoptotic and late apoptotic/necrotic cells were detected (1.8% and 1.7% for donor 1; 2.1% and 1.9% for donor 2, respectively) ( Figure 3). The good antitumor activity of compound 6 against human cancer cells and the simultaneous marginal cytotoxicity of the same compound when tested against normal human cells (PBMCs), suggest that (-)-rabdosiin may display less toxic side effects when administered in vivo. In support of our results, the few studies carried out in the last decade on the potential anticancer activity of O. sanctum extracts and its essential oil with different human cancer cell lines, clearly suggest that Tulsi may be used as a supplement to enhance anticancer chemotherapy without causing severe damage to normal epithelial cells [25,53,54]. Botanical drugs are currently approved in therapy with specific indications and in the last decades, research has focused on the anticancer effect of plant extracts.
Taken altogether, (-)-rabdosiin displays an interesting proapoptotic activity against cancer cell lines and in parallel shows a noticeable selectivity to malignant cells. It is noteworthy that the cytotoxic response of the extract is better compared to the other isolated compounds, including compound 6. As (-)-rabdosiin is a minor compound of the plant, we assume that it contributes to the improved antiproliferative activity of the methanol extract, and that it is probably synergistically with other active metabolites. The good activity of the polar extract, as well as of compound 6 against a series of human cancer cell lines and its marginal cytotoxicity against PBMCs, give evidence toward the effective use of this plant for the prevention of human cancer. Moreover, the core structure of (-)rabdosiin could be considered as drug lead in anticancer drug design.
Supplementary Material: The following are available online at www.mdpi.com/xxx/s1, Table S1: 1 Η-NMR of 6 (CD3OD, 400MHz); Table S2: 13 C-NMR of 6 (CD3OD, 400MHz); Figure S1: 1 H-NMR spectrum of 6 (CD3OD, 400 Hz); Figure S2: COSY spectrum of 6 (CD3OD, 400 Hz); Figure S3: 13 C NMR spectrum of 6 (CD3OD, 400 Hz); Figure S4: HSQC spectrum of 6 (CD3OD, 400 Hz); Figure S5: HMBC spectrum of 6 (CD3OD, 400 Hz); Figure S6: Most important HMBC signals of compound 6. The good antitumor activity of compound 6 against human cancer cells and the simultaneous marginal cytotoxicity of the same compound when tested against normal human cells (PBMCs), suggest that (-)-rabdosiin may display less toxic side effects when administered in vivo. In support of our results, the few studies carried out in the last decade on the potential anticancer activity of O. sanctum extracts and its essential oil with different human cancer cell lines, clearly suggest that Tulsi may be used as a supplement to enhance anticancer chemotherapy without causing severe damage to normal epithelial cells [25,53,54]. Botanical drugs are currently approved in therapy with specific indications and in the last decades, research has focused on the anticancer effect of plant extracts.
Taken altogether, (-)-rabdosiin displays an interesting proapoptotic activity against cancer cell lines and in parallel shows a noticeable selectivity to malignant cells. It is noteworthy that the cytotoxic response of the extract is better compared to the other isolated compounds, including compound 6. As (-)-rabdosiin is a minor compound of the plant, we assume that it contributes to the improved antiproliferative activity of the methanol extract, and that it is probably synergistically with other active metabolites. The good activity of the polar extract, as well as of compound 6 against a series of human cancer cell lines and its marginal cytotoxicity against PBMCs, give evidence toward the effective use of this plant for the prevention of human cancer. Moreover, the core structure of (-)-rabdosiin could be considered as drug lead in anticancer drug design.