**Antiproliferative Aspidosperma-Type Monoterpenoid Indole Alkaloids from** *Bousigonia mekongensis* **Inhibit Tubulin Polymerization**

**Yu Zhang 1,2, Masuo Goto 2,\*, Akifumi Oda 3, Pei-Ling Hsu 2, Ling-Li Guo 1, Yan-Hui Fu 1, Susan L. Morris-Natschke 2, Ernest Hamel 4, Kuo-Hsiung Lee 2,5,\* and Xiao-Jiang Hao 1,\***


Received: 7 February 2019; Accepted: 28 March 2019; Published: 31 March 2019

**Abstract:** Monoterpenoid indole alkaloids are structurally diverse natural products found in plants of the family Apocynaceae. Among them, vincristine and its derivatives are well known for their anticancer activity. *Bousigonia mekongensis*, a species in this family, contains various monoterpenoid indole alkaloids. In the current study, fourteen known aspidosperma-type monoterpenoid indole alkaloids (**1**–**14**) were isolated and identified from a methanol extract of the twigs and leaves of *B. mekongensis* for the first time. Among them, compounds **3**, **6**, **9**, and **13** exhibited similar antiproliferative activity spectra against A549, KB, and multidrug-resistant (MDR) KB subline KB-VIN cells with IC50 values ranging from 0.5–0.9 μM. The above alkaloids efficiently induced cell cycle arrest at the G2/M phase by inhibiting tubulin polymerization as well as mitotic bipolar spindle formation. Computer modeling studies indicated that compound **7** likely forms a hydrogen bond (H-bond) with α- or β-tubulin at the colchicine site. Evaluation of the antiproliferative effects and SAR analysis suggested that a 14,15-double bond or 3α-acetonyl group is critical for enhanced antiproliferative activity. Mechanism of action studies demonstrated for the first time that compounds **3**, **4**, **6**, **7**, and **13** efficiently induce cell cycle arrest at G2/M by inhibiting tubulin polymerization by binding to the colchicine site.

**Keywords:** aspidosperma-type; monoterpenoid indole alkaloids; antiproliferative activity; tubulin inhibitor; *Bousigonia mekongensis*

### **1. Introduction**

Microtubule-binding agents have been developed as an effective therapy in cancer treatment due to the key roles of microtubules in cell proliferation, signal transduction, and cell migration [1]. Currently, many microtubule-binding agents, including taxanes, vinca alkaloids, epothilones, halichondrins, maytansinoids, colchicine-site binding agents, and others, have been discovered from natural products

and later progressed to clinical studies and clinical use [2]. However, innate and acquired drug resistance, especially multidrug resistance (MDR), are major obstacles in cancer chemotherapy [3]. Overexpression of P-glycoprotein (P-gp) encoded by the ABCB1 gene leads to poor disease prognosis, and most clinical antimicrotubule drugs, including paclitaxel (PXL), vincristine (VIN), halichondrin B, and their analogs, are P-gp substrates [4]. A recent study indicated that antimitotic agents that target the colchicine site (CS) on the α/β-tubulin dimer were generally active in cells overexpressing βIII-tubulin, which is important in tumor aggressiveness and resistance to chemotherapy [5]. Hence, the discovery of other antimitotic CS-targeting agents might be a valuable approach for effective cancer chemotherapy, especially agents with enhanced tumor specificity and insensitivity to chemoresistance mechanisms.

Monoterpenoid indole alkaloids (MIAs) are secondary metabolites characteristic of plants in the family Apocynaceae [6]. Among them, the vinca alkaloids exhibit significant anticancer activity and vincristine, vinblastine, vinorelbine, vindesine, and vinflunine have been approved for clinical use in the treatment of hematological and lymphatic neoplasms [7]. Previous chemical studies on MIAs mostly focused on the dimeric compounds (vindoline-catharanthine), which generally exhibit superior anticancer activities compared with the corresponding monomeric units. Aspidosperma-type MIAs contain only the vindoline structural unit found in vincristine and are widely distributed in the genera *Tabernaemontana*, *Melodinus*, and *Bousigonia* (family Apocynaceae) [6]. In prior biological activity studies, several aspidosperma-type MIAs, such as jerantinines A, B, and E from *T. corymbose*, displayed significant antiproliferative activity against cancer cells [8]. Further mechanistic studies demonstrated that these alkaloids significantly arrested cells at the G2/M phase by inhibiting tubulin polymerization and, thus, they merit development as potential chemotherapeutic agents [9–11].

The genus *Bousigonia* (family Apocynaceae) contains only two species (*B. mekongensis* and *B. angustifolia*), distributed mainly in Southern China, Laos, and Vietnam [12]. Previous chemical investigation on this genus conducted in our group resulted in a series of new eburnamineaspidospermine-type bisindole alkaloids and aspidosperma-type MIAs [13–15]. As part of our ongoing work, the present study investigated the antiproliferative activity of aspidosperma-type MIAs (**1**–**14**) against five cancer cell lines, A549, MDA-MB-231, KB, P-gp-overexpressing KB subline KB-VIN, and MCF-7, as well as a primary structure activity relationship (SAR) analysis including computer modeling. The detailed mechanisms of action of these alkaloids (Figure 1) were also further investigated and are described herein.

**Figure 1.** Structures of aspidosperma-type MIAs (**1**–**14**) from *B. mekongensis*.

### **2. Results and Discussion**

### *2.1. Chemistry*

Fourteen aspidosperma-type monoterpenoid indole alkaloids, tabersonine (**1**) [16], vincadifformine (**2**) [16], 3*α*-acetonyl-tabersonine (**3**) [17], melodinine S (**4**) [18], lochnericine (**5**) [19], 14,15-*α*-epoxy-11-methoxytabersonine (**6**) [20], 11-hydroxylochnericine (**7**) [20], pachysiphine (**8**) [21], lochnerinine (**9**) [22], 19-(*R*)-hydroxytabersonine (**10**) [23], 19-(*R*)-acetoxytabersonine (**11**) [24], 11-methoxytabersonine (**12**) [16], 19-(*R*)-acetoxy-11-hydroxytabersonine (**13**) [24], 19-(*R*)-acetoxy-11 methoxytabersonine (**14**) [24] were isolated and identified from the methanol extract of *B. mekongensis* for the first time by a combination of chromatographic and spectroscopic methods (Figure 1, Tables S1–S3). The effects of the substituents at positions C-3, C-11, C-14/C-15, and C-19 were evaluated by considering the hydrophilic or hydrophobic properties.

### *2.2. Antiproliferative Activity of Compounds 1–14*

We tested the fourteen known aspidosperma-type alkaloids (**1**–**14**) against chemosensitive KB (originally isolated from epidermoid carcinoma of the nasopharynx) and P-gp over-expressing multidrug-resistant (MDR) KB subline KB-VIN, as well as three additional human cancer cell lines, A549 (lung carcinoma), MDA-MB-231 (triple negative breast cancer), and MCF-7 (estrogen receptor-positive breast cancer), using a sulforhodamine B (SRB) assay (Table 1). Vincristine (VIN), paclitaxel (PXL), and a CS agent combretastatin A-4 (CA-4) were also tested as positive controls.


**Table 1.** Antiproliferative activity and effect on tubulin assembly.

<sup>a</sup> Antiproliferative activity as IC50 values for each cell line, the concentration of compound that caused a 50% reduction relative to untreated cells determined by the SRB assay. <sup>b</sup> NT, not tested. <sup>c</sup> NA, not active (IC50 > 40 μM). <sup>d</sup> Inhibition of purified tubulin assembly, EC50 (μM) values of 50% inhibition (ITA). <sup>e</sup> Percent inhibition of 5 μM [ 3H] colchicine binding to 1 μM tubulin in the presence of 5 μM test compound (ICB).

### *2.3. Biological Activity Comparison*

While six alkaloids (**2**, **5**, **8**, **10**, **11**, and **14**) were essentially inactive, the remaining eight alkaloids (**1**, **3**, **4**, **6**, **7**, **9**, **12**, and **13**) exhibited antiproliferative activity against all cell lines tested in this study, including the MDR subline KB-VIN (Table 1). Particularly, alkaloids **3**, **6**, **9**, and **13** showed substantial potency against KB-VIN cells with IC50 values ranging from 0.5–0.7 μM. Notably, all eight active alkaloids were effective against both chemosensitive and MDR cells, as compared with well-known P-gp substrates VIN and PXL, which required 600-fold higher concentration against KB-VIN cells. The results indicate that the tested alkaloids are not substrates of P-gp and, thus, could be effective against tumors expressing the MDR phenotype.

Alkaloids **1**–**14** have the same carbon skeleton but differ in the substituents or oxidation state at various positions. Based on the antiproliferative activity data, a 14,15-double bond is critical for activity as alkaloids **2**, **5**, and **8** with a 14,15-single bond lost potency. Addition of a 3α-acetonyl group led to greatly increased potency (compare **3** versus **1**, **4** versus **5**). Potency was lost when the α-carbon of the ethyl group attached at C-19 of **1** was substituted with a hydroxyl (**10**) or acetoxy (**11**) moiety. The presence of a hydroxyl group at C-11 had a significant effect on potency (compare **7** versus **5**, **13** versus **11**), while a methoxy group at the same position led to both increased (~60-fold rise between **5** and **6**, **8** and **9**) and negligible (compare **1** versus **12**, **14** and **11**) potency.

Taken together, a 14,15-double bond or 3α-acetonyl group was required for antiproliferative activity against human cancer cell lines, including the MDR subline KB-VIN. A hydroxyl group at C-11 is necessary, while the effect of a methoxy group at C-11 was dependent on the parent skeleton. The compatibility of the synergistic groups will be considered in subsequent SAR studies.

### *2.4. Mechanisms of Action of 3, 4, 6, 7, and 13 in KB-VIN Cells*

Bioactive analogs **6** and **7** were tested for inhibitory effects on tubulin assembly as well as inhibition of [3H] colchicine binding to tubulin in a cell-free system, using highly purified bovine brain tubulin (Table 1). The results showed that alkaloid **6** strongly inhibited tubulin assembly with an EC50 (50% effective concentration for inhibiting tubulin assembly) value of 0.7 μM and inhibited the binding of colchicine to tubulin by 54%, while **7** showed moderate inhibition of tubulin assembly with an EC50 value of 4.6 μM. The data for **6** and **7** indicate that both alkaloids bind to tubulin and inhibit its assembly, and these effects are closely related to their antiproliferative activities against tumor cells. Thus, we further investigated whether the inhibition of tubulin polymerization was the major action of these alkaloids.

CA-4, a colchicine site (CS) agent, and other tubulin polymerization inhibitors induce cell cycle arrest at G2/M. Accordingly, we investigated whether the MIAs affected the cell cycle progression. KB-VIN cells were treated with compounds **3**, **4**, **6**, **7**, and **13** at their IC50 (1 × IC50) or three-fold IC50 (3 × IC50) concentration, and the cell cycle progression was analyzed by flow cytometer (Figure 2A). Expectedly, the accumulation of cells in the G2/M phase was observed in cells treated with alkaloids **3**, **4**, **6**, **7**, and **13**, as compared with CA-4 and vincristine (VIN).

To determine whether the cell cycle arrest was due to an antimicrotubule effect, cells treated with compounds were analyzed by immunocytochemistry using antibodies to α-tubulin for microtubules and mitotic spindles, Ser10-phosphorylated histone H3 (pH3) for the condensed chromatins and 4 ,6-diamidino-2-phenylindole (DAPI) for DNA. In cells treated with alkaloids **3**, **4**, **6**, **7**, and **13**, dotted tubulin aggregations without spindles were seen in the pH3-positive mitotic cells, while microtubules were undetectable in pH3-negative interphase cells (Figure 2B). These observations demonstrated that the tested alkaloids inhibited tubulin polymerization in both interphase and mitosis for bipolar spindle formation, inducing cell cycle arrest at G2/M, probably at prometaphase. In addition, alkaloid **7** was less active than **6**, which corresponded to the inhibitory activity observed in cell-based proliferation and cell-free tubulin assembly assays. Thus, we concluded that these alkaloids are tubulin polymerization inhibitors. Furthermore, the immunocytochemical data suggest that alkaloids **3, 4**, **6**, **7**, and **13** probably interact with tubulin in a biological manner similar to that of CA-4. These results agreed with those in a recent study on aspidosperma-type MIAs, such as jerantinine A, which potentially inhibit tubulin polymerization by binding to the CS on tubulin dimers [25].

Accordingly, we performed molecular docking studies to predict how alkaloid **7** binds to the tubulin dimer. An inactive alkaloid **5** was used as a comparison compound. In the tubulin binding assay described above (Table 1), CA-4 totally inhibited colchicine binding to tubulin, while **5** or **7** inhibited binding with 0 or 35% ICB value, respectively. Thus, compared with CA-4, alkaloid **5** and its hydroxyl analog **7** might bind differently to the CS. As expected, an overview of the predicted binding modes of **5** and **7** in the crystal structure of the α/β-tubulin dimer revealed considerable differences (Figures 3 and 4). The docked model of **5** showed an H-bond between the carbonyl oxygen and the side chain of Val181 on α-tubulin (αVal181), while that of **7** showed an H-bond between the C-11 hydroxyl group and the side chain of Val315 on β-tubulin (βVal315). Interestingly, in a cell-free tubulin assembly assay, compound **7** (EC50 4.6 μM) was more potent than **5** (EC50 > 40 μM). These analyses suggested that the binding mode of **5** was insufficient to inhibit tubulin assembly. The docking model also predicted that active compound **13** (IC50 0.6~6.7 μM) forms an H-bond with βAsn249, while less active compounds **4** (IC50 5.6~10.0 μM) and **11** (IC50 > 26.5 μM) form H-bonds with αThr179 and αSer178, respectively (Supplementary Figure S2). However, H-bonding with αVal181 may also be important and depends on a steric hindrance due to the parent skeleton (compare **3** versus **5**). This docking model suggested that the force of the H-bond between βVal315 or βAsn249 and the C-11 hydroxyl group of **7** might be critical for greater inhibition of tubulin assembly, which is also reflected in greater antiproliferative activity.

**Figure 2.** Mitotic defects in KB-VIN cells treated by compounds. (**A**) Vincristine-resistant subline KB-VIN cells were treated with compounds for 24 h at a concentration of one- or three-fold IC50 (1× IC50 or 3 × IC50). CA-4 at 0.2 μM was used as a colchicine-type tubulin polymerization inhibitor. Cell cycle distributions (sub-G1, G1, S, G2/M) were analyzed using flow cytometry after staining cells with propidium iodide (PI). (**B**) KB-VIN cells were treated with compounds for 24 h at a concentration of 3 × IC50. CA-4 was used at 0.2 μM. Fixed cells were stained with antibodies to α-tubulin (green) and phospho-histone H3 (pH3, red), and DAPI was used for DNA (blue). Stained cells were observed by confocal fluorescence microscope. The represented image is a projection of 15~20 optical sections acquired at 0.5~1 μm intervals. Normal mitotic spindle formation (arrow head) in control (DMSO) and dotted tubulin aggregations without spindles (**4**, **6**, **13**) or with multipolar spindles (VIN) were observed (arrows). Bar, 0.025 mm. Additional images are available in Supplementary Figure S1.

**Figure 3.** Predicted docking models for **5** and **7** binding to tubulin. Top 1 ranked docking models of **5** and **7** (sphere in 3D with gray in carbon, proton in white, oxygen in red, nitrogen in blue) in the colchicine site (CS, yellow circle) of the tubulin crystal structure (α and β tubulin heterodimer: α- (white) and β-tubulin (red)) (PDB: 1SA0) are shown as a ribbon diagram.

**Figure 4.** Predicted docking models for **5** and **7** binding in the CS. The crystal structures (PDB: 1SA0) of α- (white) and β-tubulin (red) are shown as ribbon diagrams. The distances calculated to be less than 5 Å between heavy atoms are represented by dashed lines. Docking models of compounds (gray skeleton with oxygen in red and nitrogen in blue) **5** (**A**) and **7** (**B**) in the CS are shown. Superimposition of docked compound **5** or **7** shows H-bonds with the side chain of αVal181 or βVal315, respectively. (**C**) Comparison of docking mode of **5** (green) with that of **7** (blue) in CS.

### **3. Materials and Methods**

### *3.1. General Experimental Procedures*

All chemicals and solvents were used as purchased. ESIMS data were obtained on a Finnigan MAT 90 spectrometer. NMR spectra were recorded on Bruker DRX-500 and Avance III -600 NMR spectrometers using TMS as an internal standard. All chemical shifts are reported in ppm, and apparent scalar coupling constants *J* are given in Hertz. Silica gel (300–400 mesh, Qingdao Marine Chemical Inc., Shandong, China), silica gel H (10–40 μm, Qingdao Marine Chemical Inc.), Lichroprep RP-18 gel (40–63 μm, Merck, Darmstadt, Germany), and Sephadex LH-20 (40–70 μm, Amersham Biosciences, Waltham, MA, USA) were used for column chromatography (CC). All target compounds were characterized and determined to be at least >95% pure by 1H NMR and analytical HPLC.

### *3.2. Plant Material*

The twigs and leaves of *B. mekongensis* were collected during April 2010 from Mengla County, Yunnan Province, PR China and identified by Mr. Jing-Yun Cui, Xishuangbanna Tropical Plant Garden. A voucher specimen (No. CUI20100419) has been deposited at the State Key Laboratory of Phytochemistry and Plant Resources in West China, Kunming Institute of Botany, Chinese Academy of Science (CAS).

### *3.3. Extraction and Isolation*

The dried twigs and leaves of *B. mekongensis* (12 kg) were extracted with CH3OH, the crude extract was dissolved in aqueous solution, and then the pH was adjusted to 2 by adding saturated tartaric acid with adequate stirring. The acidic mixture was defatted with petroleum ether (PE) and then extracted with CHCl3. The aqueous phase was basified to pH~10 with saturated Na2CO3 and then extracted with CHCl3 to obtain crude alkaloids. The crude alkaloids (60 g) were separated on a silica gel column (200–300 mesh; CHCl3/CH3OH, 1:0 → 0:1), yielding five major fractions (Fr 1–5). Fraction 1 (12.8 g) was chromatographed with a series of silica gel columns (CHCl3/acetone and CHCl3/CH3OH) to afford compounds **3** (21 mg) and **4** (16 mg). Fraction 2 (11.2 g) was further chromatographed on a reversed-phase C18 silica gel medium-pressure column (CH3OH/H2O, 1:1 → 1:0) to give four fractions (Fr 2A–2D). Fraction 2C (3.2 g) was separated on a silica gel column (300–400 mesh; PE/ acetone, 3:1), yielding three fractions (Fr 2C1–2C3). Compound **1** (16 mg) was separated from fraction 2C2 (760 mg) by semipreparative HPLC using a Waters XBridge C18 (10 × 250 mm, 5 μm) column with 70% CH3CN/H2O with added 0.1 *v*/*v* diethylamine. Compound **11** (36 mg) was obtained from fraction 2C3 (368 mg) by semipreparative HPLC using a Waters XBridge C18 (10 × 250 mm, 5 μm) column with 70% CH3OH/H2O with added 0.1 *v*/*v* diethylamine. Fraction 3 (9.8 g) was further chromatographed over a reversed-phase C18 silica gel medium-pressure column (CH3OH/H2O, 1:1 → 1:0) to give four fractions (Fr 3A–3D). Fraction 3A (480 mg) was separated by semipreparative HPLC using a Waters XBridge C18 (10 × 250 mm, 5 μm) column with 45% CH3OH/H2O to give compounds **5** (21 mg), **10** (4.0 mg) and **12** (28 mg). Fraction 3C (780 mg) was purified using a Sephadex LH-20 column eluted with CH3OH, followed by semipreparative HPLC using a Waters XBridge C18 (19 × 250 mm, 5 μm) column with 70% CH3OH/H2O to afford compounds **8** (22 mg) and **14** (48 mg). Fraction 4 (9.8 g) was further chromatographed over a reversed-phase C18 silica gel medium-pressure column (CH3OH/H2O, 1:1 → 1:0) to give four fractions (Fr 4A–4D). Fr 4A (518 mg) was separated by semipreparative HPLC using a Waters XBridge C18 (10 × 250 mm, 5 μm) column with 40% CH3OH/H2O to afford compounds **2** (9 mg), **6** (28 mg), **7** (20 mg), and **9** (12 mg). Fr 4B (300 mg) was further purified using a Sephadex LH-20 column eluted with CH3OH to afford **13** (35 mg).

### *3.4. Antiproliferative Activity Assay*

Antiproliferative activity was determined by the sulforhodamine B (SRB) colorimetric assay as previously described [26]. In brief, human tumor cell lines were cultured in RPMI-1640 medium containing 2 mM l-glutamine and 25 mM HEPES (Gibco), supplemented with 10% fetal bovine serum (Speciality Media), 100 μg/mL streptomycin, 100 IU/mL penicillin, and 0.25 μg/mL amphotericin B (Corning). MDR stock cells (KB-VIN) were maintained in the presence of 100 nM vincristine (VIN) (Sigma-Aldrich, Saint Louis, MI, USA).

Freshly trypsinized cell suspensions were seeded in 96-well microtiter plates at densities of 4000–11,000 cells per well (based on the cell lines) with compounds. After 72 h in culture with test compounds, cells were fixed in 10% trichloroacetic acid followed by staining with 0.04% sulforhodamine B (Sigma-Aldrich). The bound SRB was solubilized in 10 mM Tris-base and the absorbance at 515 nm was measured using a Microplate Reader (ELx800, Bio-Tek Instruments, Winooski, VT, USA) operated by Gen5 software (BioTek) after solubilizing the protein-bound dye with 10 mM Tris base. The mean IC50 is the concentration of agent that reduced cell growth by 50% compared with vehicle (DMSO) control under the experimental conditions and is the average from at least three independent experiments with duplicate samples.

The following human tumor cell lines were used in the assay: A549 (lung carcinoma), MDA-MB-231 (triplen-egative breast cancer), KB (originally isolated from epidermoid carcinoma of the nasopharynx), KB-VIN (VIN-resistant KB subline showing MDR phenotype by overexpressing P-gp), MCF-7 (estrogen receptor (ER)-positive, HER2-negative breast cancer). It should be noted that we confirmed the KB and KB-VIN cell lines were identical to AV-3 (ATCC number, CCL-21) as a HeLa (cervical carcinoma) contaminant by short tandem repeat (STR) profiling. All cell lines were obtained

from the Lineberger Comprehensive Cancer Center (UNC-CH) or from ATCC (Manassas, VA), except KB-VIN, which was a generous gift from Professor Y.-C. Cheng (Yale University). Paclitaxel was purchased from Sigma-Aldrich.

### *3.5. Tubulin Assays*

Inhibitory effects of compounds on tubulin assembly were evaluated using electrophoretically homogeneous bovine brain tubulin as described previously [27,28] using 1.0 mg/mL (10 μM) tubulin. The turbidity development as tubulin assembly was initiated by adding 0.4 mM GTP and followed for 20 min at 30 ◦C following a rapid temperature jump. Compound concentrations as EC50 values that inhibited the increase in turbidity by 50% relative to a control sample were determined. In colchicine inhibition assays, tubulin (1.0 μM) was incubated with 5.0 μM [3H] colchicine and 5.0 μM test compound at 37 ◦C for 10 min, when about 40–60% of maximum colchicine binding occurs in control samples.

### *3.6. Cell Cycle Analysis*

KB-VIN cells were seeded in 12-well plates at a density of 1 × <sup>10</sup><sup>5</sup> per well and incubated overnight. After 24 h of treatment with tested compounds at a concentration of one- or three-fold IC50, the cells were harvested and fixed in 70% EtOH at −20 ◦C overnight followed by staining with propidium iodide (PI) containing RNase (BD Pharmingen, Franklin Lakes, NJ, USA) for 30 min at 37 ◦C. The DNA contents of stained cells were analyzed by flow cytometer (LSRFortessa, BD Biosciences) controlled by FACSDiva software (BD Biosciences). Paclitaxel (PXL) was used at 3 or 6 μM. Combretastatin A-4 (CA-4) was purchased from Sigma-Aldrich and used at 0.2 μM. Experiments were repeated a minimum of two times.

### *3.7. Immunofluorescence Staining*

Immunocytochemical analysis was performed as previously described [26]. KB-VIN cells were grown on an 8-well chamber slide (Lab-Tech) for 24 h prior to treatment with the compound at a concentration of three-fold IC50. CA-4 was used at 0.2 μM (Figure 2A and Supporting Information, Supplementary Figure S1). After treatment of cells with the agent for 24 h, cells were fixed with 4% paraformaldehyde in PBS and permeabilized with 0.5% Triton X-100 in PBS. Fixed cells were labeled with mouse monoclonal antibody to α-tubulin (B5-1-2, Sigma) and rabbit IgG to Ser10-phosphorylated histone H3 (p-H3) (#06570, EMD Millipore), followed by FITC-conjugated antibody to mouse IgG (Sigma) and Alexa Fluor 549-conjugated antibody to rabbit IgG (Life Technologies). Nuclei were labeled with DAPI (Sigma). Fluorescently-labeled cells were observed using a confocal microscope (Zeiss, LSM700) with ZEN (black edition) software (Zeiss). The 15~20 optical sections acquired at 0.5~1 μm intervals were stacked and reconstructed using ZEN (black edition) software. Experiments were repeated at least twice for each compound. Final images were prepared using Adobe Photoshop CS3.

### *3.8. Computer Modeling*

GOLD 5.1 software with default settings was used to predict the three-dimensional (3D) structures of tubulin-ligand complexes [29]. The human tubulin 3D structure (TUBA1A and TUBB2B) used in this study was built from a Protein Data Bank entry (PDB ID: 1SA0) [30]. Absent hydrogen atoms in the crystal structure were added computationally by Hermes software 5.1 version. The active site radius was set to 10.0 Å, and the active site center was defined as the ligand center in 1SA0. The docking calculations used the quantum chemically-optimized ligand structures as the initial structures. Structural optimizations of ligands were performed with B3LYP/6-311+G (df, p) using Gaussian 09, Revision B.01 [31].

### **4. Conclusions**

Fourteen aspidosperma-type monoterpenoid indole alkaloids (**1**–**14**) were isolated from a methanol extract of the twigs and leaves of *Bousigonia mekongensis*. All isolates were evaluated for antiproliferative activity against five human tumor cell lines, including the MDR subline KB-VIN. Alkaloids **3**, **4**, **6**, **7**, and **13** showed significant antiproliferative effects against all five tumor cell lines. Because activity was retained against KB-VIN, the active alkaloids are not P-gp substrates and, thus, could be effective against tumors expressing the MDR phenotype. SAR studies suggested that the presence of a 14,15-double bond or 3α-acetonyl group is critical for antiproliferative activity. Furthermore, mechanistic studies revealed for the first time that the five active alkaloids cause significant arrest of the cell cycle progression at the G2/M cell cycle phase via inhibition of tubulin polymerization. In addition, the compounds interact with tubulin in a manner distinct from that of CA-4.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1420-3049/24/7/1256/ s1, Figure S1: Inhibition of tubulin polymerization by compounds, Figure S2: Predicted docking models for **3**, **4**, **11**, and **13** binding in the CS; Tables S1–S3: 1H and 13C NMR data (*δ*) for compounds **1**–**14**.

**Author Contributions:** Conceptualization, Y.Z. and M.G.; Methodology, M.G., P.-L.H. and E.H.; Software, A.O.; Formal Analysis, Y.Z. and M.G.; Resources, Y.-H.F. and L.-L.G.; Data Curation, Y.Z. and M.G.; Writing—Original Draft Preparation, Y.Z.; Writing—Review & Editing, M.G., S.L.M.-N. and K.-H.L.; Supervision, K.-H.L. and X.-J.H.; Project Administration, Y.Z. and X.-J.H.; Funding Acquisition, Y.Z.

**Funding:** This research was financially supported by the National Natural Science Foundation of China (No. 81874295) to Y.Z., Yunnan Applied Basic Research Projects (No. 2018FA049) to Y.Z., CAS "Light of West China" Program to Y.Z., the Youth Innovation Promotion Association of CAS (2015323), and the Young Academic and Technical Leader Raising Foundation of Yunnan Province (to Y.Z.). This study was also supported in part by NIH Grant CA177584 from the National Cancer Institute awarded to KH Lee and the Eshelman Institute for Innovation, Chapel Hill, North Carolina, awarded to M.G.

**Acknowledgments:** We wish to thank the Microscopy Service Laboratory (UNC-CH) for its expertise in the confocal microscopy studies.

**Disclaimer:** The content of this paper is solely the responsibility of the authors and does not necessarily reflect the official views of the National Institutes of Health.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


**Sample Availability:** Samples of the compounds **1**–**14** are available from the authors.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Synthesis and Biological Evaluation of Phaeosphaeride A Derivatives as Antitumor Agents**

### **Victoria Abzianidze 1,\*, Petr Beltyukov 2, Sofya Zakharenkova 1, Natalia Moiseeva 3, Jennifer Mejia 4, Alvin Holder 4, Yuri Trishin 5, Alexander Berestetskiy <sup>6</sup> and Victor Kuznetsov <sup>1</sup>**


Received: 19 October 2018; Accepted: 16 November 2018; Published: 21 November 2018

**Abstract:** New derivatives of phaeosphaeride A (PPA) were synthesized and characterized. Anti-tumor activity studies were carried out on the HCT-116, PC3, MCF-7, A549, K562, NCI-H929, Jurkat, THP-1, RPMI8228 tumor cell lines, and on the HEF cell line. All of the compounds synthesized were found to have better efficacy than PPA towards the tumor cell lines mentioned. Compound **6** was potent against six cancer cell lines, HCT-116, PC-3, K562, NCI-H929, Jurkat, and RPMI8226, showing a 47, 13.5, 16, 4, 1.5, and 7-fold increase in anticancer activity comparative to those of etoposide, respectively. Compound **1** possessed selectivity toward the NCI-H929 cell line (IC50 = 1.35 ± 0.69 μM), while product **7** was selective against three cancer cell lines, HCT-116, MCF-7, and NCI-H929, each having IC50 values of 1.65 μM, 1.80 μM and 2.00 μM, respectively.

**Keywords:** natural phaeosphaeride A; antitumor activity; human tumor cell lines; HEF cell line; acute toxicity

### **1. Introduction**

Modern chemotherapeutic treatment of malignant tumors is widely considered to have begun approximately 75 years ago when researchers found that nitrogen mustard displayed anti-tumor activity [1]. In more recent decades, leading research approaches have focused on the development of new antitumor agents for targeted therapy, as well as combined treatment with immunotherapeutic and chemotherapeutic agents [2]. Diversity, and in many cases, the genetic uniqueness of tumors, does not allow for the development of universal and specific therapeutic treatment. As a result, the use of non-specific chemotherapeutic treatment is still the principal therapeutic option used in patients. Despite advances in cancer treatments, the disabling side effects and relative effectivity of these broad antitumor drugs plagues scientists to urgently develop improved chemotherapeutic agents.

Many chemotherapeutic agents used in clinical practice are developed from natural compounds or their derivatives and analogs, e.g., etoposide, eribulin, paclitaxel, vincristine, vinblastine, topotecan, cytarabine, doxorubicin, dactinomycin, and bleomycin [3–5]. These also include medicines developed and introduced into clinical practice from the 1950s–1960s (vincristine, vinblastine, cytarabine, etc.), as well as modern agents developed in the 2000's (topotecan, eribulin). Many chemotherapeutic agents based on natural analogs continue to undergo clinical studies to evaluate their effectiveness in the treatment of various types of tumors, as drugs developed from natural compounds often show lower toxicity and have higher target values towards malignant tumor cells [3]. While scientists continue to research these compounds, these naturally derived drugs are already relevant in the market today, with close to 40% of antitumor drugs approved by the FDA being of natural origin or semi-synthetic derivatives of natural compounds. When we reexamine this number and consider only the use of synthetic drugs, the analogues of natural compounds, the proportion of these antitumor agents are estimated at 70% [6]. Clearly, nature remains an inexhaustible source of new substances and has vast potential to aid in the fight against various tumors.

Here, phaeosphaeride A, produced by endophytic fungi from the genus *Phaeosphaeria*, was chosen for its ability to inhibit the Signal Transducer and Activator of Transcription 3 (STAT3) signaling pathway. The stereochemical configuration of this natural product had been established by total synthesis of ent-phaeosphaeride A and phaeosphaeride A [7,8] and by X-ray diffraction [9]. High incidence of STAT3 protein is characteristic of several oncological diseases like leukemia, multiple myeloma, cancers of the breast and lung, as well as multiple carcinomas such as renal, prostate, hepatocellular, ovarian, and pancreatic [10,11]. STAT3 also is shown to play an important role in regulating cell growth and viability [10,12–15]. Therefore, phaeosphaeride A, and its derivatives, are potentially promising anticancer agents for targeted therapy and combined treatments [10,11].

The results of the study in this article include information on the methods for the synthesis of the phaeosphaeride A derivatives and their biological evaluation, including the measurements of the cytotoxic effects and a preliminary assessment of acute toxicity in mice.

### **2. Results and Discussion**

### *2.1. Chemistry*

The synthesis of target compounds **1**–**8** is presented in Scheme 1. Mesylation of PPA with MeSO2Cl and Et3N in CH2Cl2 gave the mesylate as a sole product, which was used in the next step without purification.

**Scheme 1.** Synthesis of compounds **1**–**8**. *Reagents and conditions*: (**a**) MsCl, TEA, CH2Cl2, 0 ◦C, 1 h; (**b**) cyclic or primary amine, with or without TEA, acetonitrile or THF, room temperature or 70–75 ◦C.

Treatment of the mesylate with cyclic amines and (or without) TEA in acetonitrile (or THF) at room temperature (or at 70–75 ◦C) or with primary amines in acetonitrile at room temperature gave the corresponding amino derivatives in 11–27% yield with inversion occurring on the C-6 atom. The ROESY spectra of the products showed a correlation between the methyl protons (H-15) and the H-6 proton confirming the inversion of configuration of the C-6 atom (see Supplementary Material). Product **6** was synthesized by our research group in 2017 with the present method doubling the total yield [16]. Previously it was impossible to obtain the desired derivatives through chloroacetyl PPA derivative from primary amines as those reactions proceeded solely at the exocyclic double bond (unpublished data) [17].

### *2.2. Biological Evaluation*

### 2.2.1. Cytotoxicity Assay Using 9 Tumor Cell Lines and HEF Cell Line

All newly synthesized PPA derivatives **1**–**8** were evaluated for their anti-proliferative activity against human breast cancer MCF-7, human prostate adenocarcinoma PC-3, human colorectal cancer HCT-116, human lung cancer A549, human chronic myelogenous leukemia K562, human acute monocytic leukemia THP-1, human multiple myeloma RPMI8226, human acute T-cell leukemia Jurkat, human multiple myeloma NCI-H929, and human embryonic fibroblasts HEF cell lines by MTT assays. All cells were incubated with different concentrations of PPA derivatives for 72 h, with etoposide and PPA used as reference compounds. The anticancer activity of the tested compounds was described as the concentration of drug inhibiting 50% cell growth IC50 (Tables 1 and 2).

**Table 1.** IC50 values for the respective compounds when studied on the adhesive cell lines. Data was expressed as the inhibitory ratio ± SD based on three independent experiments (*n* = 3).


**Table 2.** IC50 values for the respective compounds when studied on the suspension cell lines. Data were expressed as inhibitory ratio ± SD based on three independent experiments (*n* = 3).


Compound **6** was found to be the most potent against six cancer cell lines (0.47 μM for HCT-116, 0.2 μM for PC-3, 0.54 μM for K562, 0.23 μM for NCI-H929, 0.55 μM for Jurkat and 0.63 μM for

RPMI8226), which was 51, 160, 37, 28, 17.5, and 14.5-fold stronger than those of the reference compound PPA, respectively (Figure 1).

**Figure 1.** (**a**) The in vitro effects of compound **6** on the cell viability of human HCT-116 cell line; (**b**) The in vitro effects of compound **6** on the cell viability of human NCI-H929 cell line.

Compound **1** showed selectivity toward the NCI-H929 cell line (IC50 = 1.35 ± 0.69 μM) while product **7** was selective against three cancer cell lines (IC50 was 1.65 μM, 1.80 μM and 2.00 μM towards HCT-116, MCF-7 and NCI-H929, respectively). The results obtained indicate product **6** to be more toxic than the positive control, etoposide, against HCT-116 (47-fold), PC-3 (13.5-fold), K562 (16-fold), NCI-H929 (4-fold), Jurkat (1.5-fold) and RPMI8226 (7-fold) cancer cell lines. Human embryonic fibroblasts were used as a control.

### 2.2.2. Acute Intraperitoneal Toxicity Study

No adverse effects were found from compound **6** on body weight and food consumption. There was no indication of morbidity, mortality, or lethal effects during the 14 days after i.p. administration (intraperiotenal) in mice of two low doses (8.25 and 82.5 mg/kg). LD50 was not reached in all experiments. Mortality (two of eight mice) was observed in the highest dose group (one was found deceased on Day 2 and one on Day 3). No effect was observed on mean body weight or mass coefficients (spleen, liver, heart, kidney, lung, thymus, and adrenal). No gross abnormalities were observed in the organs tested. Maximum tolerated dose of **6** after i.p. injection in mice was equal to or more than 82.5 mg/kg. Median lethal dose (LD50) of **6** has not been determined but is believed to be more than 200 mg/kg. It can be concluded that compound **6** is less toxic comparative to etoposide. Intraperitoneal LD50 of etoposide for mice is 64 mg/kg (RTECS # KC0190000).

### **3. Materials and Methods**

### *3.1. Materials and Instruments*

1H-NMR spectra were acquired on an AVANCE III 400 MHz NMR spectrometer (Bruker, Rheinstetten, Germany) in CDCl3. Optical rotations were acquired on a Polaar 3005 Polarimeter (Optical Activity, Huntingdon, Great Britain) using a 2.5 cm cell with a Na 589 nm filter and the concentration of samples was denoted as *c*. Mass spectra data were acquired on a TSQ Quantum Access Max Mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). High-resolution mass spectra (HRMS) were acquired on a LTQ Orbitrap Velos spectrometer (Thermo Scientific) and on a Bruker MicrOTOF. FTIR spectra were acquired on an IR Affinity-1 spectrometer (Shimadzu, Thermo Scientific). Organic solvents used were dried by standard methods when necessary. Commercially available reagents were used without further purification. All reactions were monitored by TLC with silica gel coated plates (EMD/Merck KGaA, Darmstadt, Germany), with visualization by UV light and by charring with 0.1% ninhydrin in EtOH. Column chromatography was performed using Merck 60 Å 70–230 mesh silica gel. The optical density was determined using a Multiskan FC spectrophotometer (Thermo Scientific) at a wavelength of 540 nm when using the MTT assay.

### *3.2. Chemical Syntheses*

3.2.1. Synthesis of (2*S*,3*S*,4*S*)-3-Hydroxy-6-methoxy-3-methyl-7-methylene-5-oxo-2-pentyl-2,3,4,5,6,7-hexahydropyrano[2,3-*c*]pyrrol-4-yl methanesulfonate

PPA (1 mmol) was dissolved in CH2Cl2 (2 mL) and cooled to 0 ◦C. Triethylamine (3.5 mmol) was added followed by the dropwise addition of methanesulfonyl chloride (2.5 mmol). The mixture was stirred at 0 ◦C for 1 h. The reaction was quenched by the addition of saturated NaHCO3 solution and the mixture was extracted with dichloromethane (3 × 20 mL). The organic layer was washed with water and brine, dried over magnesium sulfate, and concentrated in vacuo after filtration. Mesylate was used in the next step without further purification. HRMS revealed an [M + H]+ion with exact mass 376.14219, corresponding to the molecular formula C16H26NO7S.

### 3.2.2. General Procedure for the Synthesis of Compounds **1**–**5**

A mixture of the crude (2*S*,3*S*,4*S*)-3-hydroxy-6-methoxy-3-methyl-7-methylene-5-oxo-2-pentyl-2,3,4,5,6,7-hexahydropyrano[2,3-*c*]pyrrol-4-yl methanesulfonate (1 mmol) and an appropriate primary amine (2 mmol) was stirred in dry acetonitrile (2 mL) at room temperature until consumption of the starting material was complete as judged by TLC analysis (24–48 h). The reaction mixture was diluted with ether (20 mL) and transferred to a separatory funnel. The layers were separated and the aqueous layer was extracted with ether (3 × 20 mL). The organic extracts were combined, washed with brine (2 × 20 mL), dried over magnesium sulfate, and concentrated in vacuo. The crude product was purified by flash chromatography (DCM:methanol = 60:1).

*(2S,3R,4R)-3-Hydroxy-6-methoxy-3-methyl-4-(methylamino)-7-methylene-2-pentyl-3,4,6,7-tetrahydro-pyrano [2,3-c]pyrrol-5(2H)-one* (**1**): Yield 17%, yellow oil. [α] 20.0 *<sup>D</sup>* <sup>=</sup> −120.36 (c 0.28, CH2Cl2). 1H-NMR (CDCl3) δ 5.56 (m, 2H), 5.12 (s, 1H), 5.07 (s, 1H), 4.05 (m, 1H), 3.93 (s, 3H), 3.41 (m, 1H), 2.75 (s, 3H), 1.93 (m, 1H), 1.63 (m, 2H), 1.43–1.33 (m, 5H), 1.15 (s, 3H), 0.90 (m, 3H). 13C-NMR (CDCl3) δ 165.87 (s), 159.47 (s), 136.05 (s), 99.07 (s), 93.27 (s), 81.54 (s), 68.97 (s), 64.70 (s), 57.32 (s), 33.50 (s), 31.58 (s), 27.62 (s), 25.99 (s), 22.53 (s), 19.45 (s), 14.05 (s). IR (KBr) 3319, 2955, 2930, 2859, 1721, 1667, 1633, 1438, 1379, 1266, 1196, 1166, 1085, 979, 915 cm<sup>−</sup>1. HRMS [M + H]+ calcd for C16H27N2O4 311.19653, found 311.1953.

*(2S,3R,4R)-3-Hydroxy-4-[(2-hydroxyethyl)amino]-6-methoxy-3-methyl-7-methylene-2-pentyl-3,4,6,7-tetrahydropyrano[2,3-c]pyrrol-5(2H)-one* (**2**): Yield 25%, yellow oil. [α] 20.0 *<sup>D</sup>* <sup>=</sup> −99.54 (c 0.22, CH2Cl2). 1H-NMR (CDCl3) δ 5.09 (s, 1H), 5.04 (s, 1H), 3.93 (m, 4H), 3.61 (m, 2H), 3.12 (s, 1H), 3.06–2.91 (m, 2H), 1.94 (m, 1H), 1.65–1.58 (m, 2H), 1.35 (m, 6H), 1.06 (s, 3H), 0.91 (s, 3H). 13C-NMR (CDCl3) δ 167.38 (s), 157.36 (s), 136.27 (s), 104.10 (s), 92.86 (s), 82.53 (s), 69.28 (s), 64.71 (s), 61.07 (s), 54.14 (s), 52.07 (s), 31.66 (s), 27.47 (s), 26.28 (s), 22.57 (s), 18.58 (s), 14.08 (s). IR (KBr) 3322, 2954, 2927, 2857, 1718, 1633, 1547, 1458, 1437, 1378, 1263, 1191, 1117, 1081, 978, 914 cm−1. HRMS [M + H]+ calcd for C17H29N2O5 341.20710, found 341.2060.

*(2S,3R,4R)-3-Hydroxy-4-[(4-hydroxybutyl)amino]-6-methoxy-3-methyl-7-methylene-2-pentyl-3,4,6,7-tetrahydropyrano[2,3-c]pyrrol-5(2H)-one* (**3**): Yield 20%, yellow oil. [α] 20.0 *<sup>D</sup>* <sup>=</sup> −89.80 (c 0.51, CH2Cl2). 1H-NMR (CDCl3) δ 5.11 (d, *J* = 1.2 Hz, 1H), 5.05 (d, *J* = 1.3 Hz, 1H), 4.94 (m, 3H), 4.05 (m, 1H), 3.93 (s, 3H), 3.80–3.50 (m, 2H), 3.36 (s, 1H), 3.27–3.12 (m, 1H), 3.06–2.87 (m, 1H), 1.95 (m, 1H), 1.78 (m, 2H), 1.69–1.55 (m, 4H), 1.45–1.33 (m, 5H), 1.12 (s, 3H), 0.90 (m, 3H). 13C-NMR (CDCl3) δ 166.37 (s), 158.18 (s), 136.38 (s), 102.16 (s), 92.59 (s), 82.14 (s), 68.78 (s), 64.64 (s), 62.26 (s), 55.97 (s), 48.77 (s), 31.64 (s), 30.14 (s), 27.61 (s), 26.12 (s), 22.57 (s), 19.15 (s), 14.08 (s). IR (KBr) 3292, 2927, 2858, 1712, 1632, 1548, 1438, 1379, 1191, 1083, 989, 914 cm<sup>−</sup>1. HRMS [M + H]+ calcd for C19H33N2O5 369.23840, found 369.2373.

*(2S,3R,4R)-4-(Allylamino)-3-hydroxy-6-methoxy-3-methyl-7-methylene-2-pentyl-3,4,6,7-tetrahydropyrano- [2,3-c]pyrrol-5(2H)-one* (**4**): Yield 14%, yellow oil. [α] 20.0 *<sup>D</sup>* <sup>=</sup> −157.50 (c 0.31, CH2Cl2). 1H-NMR (CDCl3) <sup>δ</sup>

5.89 (ddt, *J* = 12.2, 10.1, 6.1 Hz, 1H), 5.25 (dd, *J* = 17.2, 1.3 Hz, 1H), 5.12 (m, 1H), 5.03 (d, *J* = 1.3 Hz, 1H), 4.99 (d, *J* = 1.3 Hz, 1H), 3.93 (s, 3H), 3.78–3.66 (m, 1H), 3.58–3.49 (m, 1H), 3.44 (ddd, *J* = 13.8, 4.6, 1.2 Hz, 1H), 3.03 (s, 1H), 1.94 (m, 1H), 1.60 (m, 2H), 1.43–1.35 (m, 5H), 1.06 (s, 3H), 0.91 (m, 3H). 13C-NMR (CDCl3) δ 166.68 (s), 156.86 (s), 136.78 (s), 136.06 (s), 117.05 (s), 105.17 (s), 91.79 (s), 82.65 (s), 68.23 (s), 64.59 (s), 54.92 (s), 52.14 (s), 31.69 (s), 27.63 (s), 26.29 (s), 22.57 (s), 18.72 (s), 14.09 (s). IR (KBr) 3314, 2955, 2928, 2858, 1720, 1668, 1634, 1437, 1367, 1232, 1192, 1087, 992, 919 cm−1. HRMS [M + H]+ calcd for C18H29N2O4 337.21218, found 337.2111.

*(2S,3R,4R)-4-(benzylamino)-3-hydroxy-6-methoxy-3-methyl-7-methylene-2-pentyl-3,4,6,7-tetrahydro-pyrano [2,3-c]pyrrol-5(2H)-one* (**5**): Yield 17%, yellow oil. [α] 20.0 *<sup>D</sup>* <sup>=</sup> −137.81 (c 0.43, CH2Cl2). 1H-NMR (CDCl3) <sup>δ</sup> 7.39–7.26 (m, 5H), 5.04 (d, *J* = 1.4 Hz, 1H), 5.00 (d, *J* = 1.4 Hz, 1H), 4.30 (d, *J* = 12.3 Hz, 1H), 3.94 (m, 4H), 3.53 (m, 1H), 3.12 (s, 1H), 1.93 (m, 1H), 1.61 (m, 2H), 1.34 (m, 5H), 1.06 (s, 3H), 0.90 (m, 3H). 13C-NMR (CDCl3) δ 166.78 (s), 156.89 (s), 139.35 (s), 136.81 (s), 128.61 (d, *J* = 8.7 Hz), 127.47 (s), 105.30 (s), 91.89 (s), 82.70 (s), 68.26 (s), 64.63 (s), 55.37 (s), 53.62 (s), 31.68 (s), 27.63 (s), 26.27 (s), 22.56 (s), 18.82 (s), 14.07 (s). IR (KBr) 3309, 2955, 2927, 2858, 1720, 1633, 1454, 1436, 1378, 1232, 1192, 1086, 978, 914 cm<sup>−</sup>1. HRMS [M + H]+ calcd for C22H31N2O4 387.22783, found 387.2262.

### 3.2.3. General Procedure for the Synthesis of Compounds **6**–**8**

A mixture of the crude (2*S*,3*S*,4*S*)-3-hydroxy-6-methoxy-3-methyl-7-methylene-5-oxo-2-pentyl-2,3,4,5,6,7-hexahydropyrano[2,3-*c*]pyrrol-4-yl methanesulfonate (1 mmol), the corresponding cyclic amine (1.5 mmol) and triethylamine (3 mmol) was stirred in dry acetonitrile (2 mL) at room temperature until consumption of the starting material was complete as judged by TLC analysis (24–48 h). The reaction mixture was quenched with water and extracted with EtOAc (2 × 20 mL). The organic extract was washed with brine, dried over magnesium sulfate, and concentrated in vacuo. The crude product was purified by flash chromatography (DCM/methanol).

### 3.2.4. Alternative General Procedure for the Synthesis of Compounds **6**–**8**

A mixture of the crude (2*S*,3*S*,4*S*)-3-hydroxy-6-methoxy-3-methyl-7-methylene-5-oxo-2-pentyl-2,3,4,5,6,7-hexahydropyrano[2,3-*c*]pyrrol-4-yl methanesulfonate (1 mmol) and the corresponding cyclic amine (5 mmol) in dry THF (2 mL) was heated at 70–75 ◦C in a sealed tube for 18 h. The mixture was cooled to room temperature, filtered, and the THF solution was diluted with saturated sodium bicarbonate solution (20 mL). The resulting mixture was extracted with DCM (3 × 20 mL). The organic extracts were washed with water, dried with magnesium sulfate, and concentrated in vacuo. The crude product was purified by flash chromatography (DCM/methanol).

*(2S,3R,4R)-3-Hydroxy-6-methoxy-3-methyl-7-methylene-2-pentyl-4-pyrrolidin-1-yl-3,4,6,7-tetrahydro-pyrano [2,3-c]pyrrol-5(2H)-one* (**6**): Yield 27%, yellow oil. [α] 20.0 *<sup>D</sup>* <sup>=</sup> −176.52 (c 0.40, CH2Cl2). 1H-NMR (CDCl3) δ 5.40 (s, 1H), 5.02 (d, *J* = 1.4 Hz, 1H), 4.98 (d, *J* = 1.4 Hz, 1H), 3.92 (s, 3H), 3.64 (d, *J* = 9.9 Hz, 1H), 3.31 (s, 1H), 3.10–2.55 (m, 4H), 2.03–1.89 (m, 1H), 1.84–1.70 (m, 4H), 1.69–1.52 (m, 2H), 1.45–1.29 (m, 5H), 1.05 (s, 3H), 0.91 (t, *J* = 6.7 Hz, 3H). 13C-NMR (CDCl3) δ 167.11 (s), 158.44 (s), 137.00 (s), 101.65 (s), 91.38 (s), 83.60 (s), 68.00 (s), 64.41 (s), 59.55 (s), 31.73 (s), 28.21 (s), 26.43 (s), 23.91 (s), 22.55 (s), 19.66 (s), 14.04 (s). IR (KBr) 3430, 2957, 2930, 2860, 1722, 1633, 1438, 1190, 1144 cm−1. HRMS [M + H]+, calcd. for C19H31N2O4 351.227834, found 351.22733.

*(2S,3R,4R)-3-Hydroxy-6-methoxy-3-methyl-7-methylene-2-pentyl-4-(4-pyrrolidin-1-ylpiperidin-1-yl)-3,4,6,7 tetrahydropyrano[2,3-c]pyrrol-5(2H)-one* (**7**): Yield 11%, yellow oil. [α] 20.0 *<sup>D</sup>* <sup>=</sup> <sup>−</sup>115.82 (c 0.41, CH2Cl2). 1H-NMR (CDCl3) <sup>δ</sup> 5.05 (s, 1H), 5.00 (s, 1H), 3.91 (s, 3H), 3.56 (m, 1H), 3.29–2.91 (m, 8H), 2.81 (m, 1H), 2.65 (s, 1H), 2.27 (t, *J* = 10.9 Hz, 1H), 2.11–1.83 (m, 9H), 1.69–1.48 (m, 2H), 1.35 (s, 5H), 1.02 (s, 3H), 0.91 (s, 3H). 13C-NMR (CDCl3) δ 167.08 (s), 158.19 (s), 136.87 (s), 101.35 (s), 91.69 (s), 83.39 (s), 68.22 (s), 64.49 (s), 62.51 (s), 61.35 (s), 55.67 (s), 51.39 (s), 49.52 (s), 32.19 (s), 31.74 (s), 28.29 (s), 26.44 (s), 23.20 (s), 22.57 (s), 19.96 (s), 14.08 (s). IR (KBr) 3213, 2928, 2857, 2782, 1724, 1669, 1635, 1437, 1377, 1354, 1321, 1193, 1150, 1126, 1085, 979 cm<sup>−</sup>1. HRMS [M + H]+ calcd for C24H40N3O4 434.30133, found 434.3009.

*(2S,3R,4R)-3-Hydroxy-4-(4-hydroxypiperidin-1-yl)-6-methoxy-3-methyl-7-methylene-2-pentyl-3,4,6,7-tetrahydropyrano[2,3-c]pyrrol-5(2H)-one* (**8**): Yield 16%, yellow oil. [α] 20.0 *<sup>D</sup>* <sup>=</sup> <sup>−</sup>111.11 (c 0.36, CH2Cl2). 1H-NMR (CDCl3) <sup>δ</sup> 5.17 (s, 1H), 5.03 (d, *<sup>J</sup>* = 1.1 Hz, 1H), 5.00 (d, *<sup>J</sup>* = 1.1 Hz, 1H), 3.93 (s, 3H), 3.68 (s, 1H), 3.59 (m, 1H), 3.02 (s, 1H), 2.87 (s, 2H), 2.35 (s, 1H), 2.03–1.85 (m, 3H), 1.68–1.50 (m, 6H), 1.35 (m, 5H), 1.05 (s, 3H), 0.91 (m, 3H). 13C-NMR (CDCl3) δ 167.01 (s), 158.15 (s), 136.85 (s), 101.35 (s), 91.75 (s), 83.35 (s), 68.20 (s), 64.53 (s), 62.45 (s), 35.20 (s), 31.79 (s), 28.35 (s), 26.48 (s), 22.58 (s), 20.00 (s), 14.09 (s). IR (KBr) 3213, 2926, 2855, 1720, 1671, 1634, 1547, 1438, 1375, 1139, 1066, 1045 cm−1. HRMS [M + H]+ calcd for C20H33N2O5 381.23840, found 381.2377.

### *3.3. Bio-Evaluation Methods*

### 3.3.1. In-Vitro Cytotoxicity Study (MTT Assay)

MCF-7 (breast cancer), PC-3 (prostate adenocarcinoma), HCT-116 (colorectal cancer cell), A549 (lung cancer), K562 (chronic myelogenous leukemia), THP-1 (acute monocytic leukemia), and RPMI8226 (multiple myeloma), Jurkat (acute T-cell leukemia), HEF (human embryonic fibroblasts) was purchased from the Russian Academy of Sciences Cells Bank (Institute of Cytology of the Russian Academy of Sciences, Saint Petersburg, Russian Federation). Multiple myeloma cell line NCI-H929 was purchased from ATCC (USA). The MCF-7, A549 and HEF (human embryonic fibroblasts) were cultured in DMEM medium. Other cell lines were cultured in RPMI1640 medium, (PanEco, Russia) supplemented with 10% fetal bovine serum (GE Healthcare LifeSciences, São Paulo, Brazil), and gentamicin at a concentration of 40 μg/mL and cultured at 37 ◦C in a humidified atmosphere containing 5% CO2. All experiments were performed with cells at passages 3 to 7 in the logarithmic phase of growth.

Cells were seeded into 96-well plates of 5 × <sup>10</sup><sup>3</sup> for adhesive cultures and 20 × <sup>10</sup><sup>3</sup> for suspension cultures per well in 90 μL and 135 μL of culture medium, respectively. The test substances were dissolved in DMSO to a concentration of 1 × <sup>10</sup>−<sup>2</sup> M. For subsequent dilution of the substances, a serum-free culture medium was used as the diluent. Final concentration of DMSO in wells was no more than 1%. It was found that this concentration of DMSO didn't affect the cells. Substances were added to the cells in 3–4 replicates after 24 h for adhesive cultures and immediately for suspension cultures. 10–15 μL of serum-free medium was added to the control wells with non-exposed cells. Cells were cultured for 72 h at 37 ◦C in an atmosphere of 5% CO2.

### 3.3.2. Experimental Animals

Six to eight-week-old outbreed male mice were purchased from the Rappolovo Animal Farm of the Russian Academy of Sciences. Animals were group housed in solid bottom polycarbonate cages (3–5 animals/cage) and provided with sterilized pelleted food and pure water ad libitum.

### 3.3.3. Ethical Guidelines

Animal care and protocols of the study were conducted in compliance with ethical standards and recommendations for the human treatment of animals used in experimental and other scientific purposes according to the European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes (and protocol of amendment ETS No 170), the National Standard of the Russian Federation GOST R-53434-2009, "Principles of Good Laboratory Practice," and by the order of the Ministry of Health of the Russian Federation 01.04.2016, No 199n «On approval of the rules of good laboratory practice».

### 3.3.4. Acute Intraperitoneal Toxicity Study

Thirty mice were randomly assigned to the acute toxicity study. Animals in experimental groups (8 per/group) received a dose formulation containing compound **6** solution in 5% DMSO at various dosages (8.25, 82.5 and 200 mg/kg) via single i.p. injection. Mice in the control group treated with vehicle (5% DMSO). The location of the i.p. injection was in the lower left abdominal quadrant. Volume of injection did not exceed 250 μL. After treatments, the following parameters and end points were evaluated for 14 days: mortality, clinical signs, body weight, food consumption, locomotion, salivation, diarrhea, and lethargy. The maximum tolerated dose in this study is defined as the highest dose that will be tolerated and not produce major life-threatening toxicity for the study duration [18]. All experimental procedures were performed according to the principles and guidelines for the care and use of laboratory animals. The animals were sacrificed by carbon dioxide asphyxiation and cervical dislocation. All efforts were made to minimize suffering. Post mortem evaluation included gross examination for all animals at terminal necropsy and calculation of organ mass coefficients. No histopathological examinations were performed.

### **4. Conclusions**

In summary, a series of PPA derivatives were synthesized via mesylation and amination. Some of these derivatives represent a promising class of cytotoxic agents with potential therapeutic values. Compound **6** demonstrated the highest cytotoxicity and was less toxic when used in vivo compared with the clinically used etoposide.

**Supplementary Materials:** 1H-NMR, 13C-NMR, 1H-1H ROESY and HRMS of these compounds are available in the supplementary materials.

**Author Contributions:** V.A. performed the research, analyzed the data and wrote the paper. S.Z. carried out the synthesis of the target compounds. P.B. and N.M. performed the biological assay of the products. J.M. and A.H. analyzed data and revised the manuscript. Y.T. and V.K. conceived the work, gave critical comments. P.B. performed animal studies. A.B. isolated phaeosphaeride A from a fungal strain.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


**Sample Availability:** Samples of compounds **1**–**8** are available from the authors.

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Review* **Anticancer Potential of Resveratrol,** β**-Lapachone and Their Analogues**

**Danielly C. Ferraz da Costa 1,**†**, Luciana Pereira Rangel 2,**†**, Mafalda Maria Duarte da Cunha Martins-Dinis 3,**†**, Giulia Diniz da Silva Ferretti 3,**†**, Vitor F. Ferreira <sup>4</sup> and Jerson L. Silva 3,\***


Received: 31 December 2019; Accepted: 13 February 2020; Published: 18 February 2020

**Abstract:** This review aims to explore the potential of resveratrol, a polyphenol stilbene, and beta-lapachone, a naphthoquinone, as well as their derivatives, in the development of new drug candidates for cancer. A brief history of these compounds is reviewed along with their potential effects and mechanisms of action and the most recent attempts to improve their bioavailability and potency against different types of cancer.

**Keywords:** resveratrol; β-lapachone; cancer

### **1. Introduction**

Cancer is a critical public health problem worldwide, with more than 18 million new cases and 9.6 million deaths estimated in 2018 [1]. Cancer therapeutics involves multiple combined approaches and requires the development of strategies based on the design and synthesis of promising compounds to improve treatment response. The search for new molecules with antitumor activity is still necessary and pursued by the pharmaceutical industry and many different research groups. Natural compounds have long been used for this purpose, leading to the development of new drugs, or as templates for new molecules with similar structures and effects [2]. Historically, bioactive compounds derived from animals and plants have been extensively used to treat diseases, which explain the scientific interest in natural products for drug discovery [3]. Both resveratrol and β-lapachone have been used with this purpose, leading to the development of new derivatives and the production of delivery systems aimed at improving the bioavailability of these compounds. In this review, we describe the potential of resveratrol, a polyphenol stilbene, and β-lapachone, a napthoquinone (Figure 1), as well as their derivatives, in the development of new drug candidates for cancer. We begin with a brief history of these compounds and move further to the discussion of their potential effects and mechanisms of action and the most recent attempts to improve their bioavailability and potency against different types of cancer.

**Figure 1.** Structures of resveratrol and β-lapachone.

### **2. Resveratrol and Stilbene-Based Compounds**

Stilbenes are phytochemicals with small molecular weights (approximately 200–300 g/mol) found in a wide range of plants and dietary supplements [4]. Stilbene-based compounds have become of particular interest because of their wide range of biological activities. Among them, resveratrol (3,4 ,5-trihydroxy-*trans*-stilbene) is a natural nonflavonoid polyphenol, classified as a phytoalexin, which can be naturally produced by more than 70 plant species (including grapes, blueberries, raspberries, mulberries and peanuts) in response to stressful conditions, such as fungal infection and ultraviolet radiation. In plants, the molecule exists as *trans*-resveratrol and *cis*-resveratrol isomers, and their glucosides, *trans*-piceid and *cis*-piceid. Since resveratrol is efficiently extracted from grape skin during the wine-making process, red wine is the most important dietary source of this bioactive compound [5–7]. Resveratrol has attracted scientific attention since 1997, when Jang et al. first demonstrated its ability to modulate in vivo carcinogenesis by inhibiting tumor initiation, promotion and progression in mice [8]. After that, the number of published papers regarding the role of resveratrol in blocking the multistep process of carcinogenesis increased substantially [9,10].

Currently, resveratrol is well characterized as a potent chemopreventive and chemotherapeutic agent in different cancer experimental models and clinical trials (for a review, see our recent publication) [11]. Several cell processes are targeted by this phytoalexin by upregulation or downregulation of multiple molecular pathways involved in cancer. Resveratrol modulates xenobiotic metabolism by inhibiting phase I cytochrome P450 enzymes responsible for carcinogen activation and by inducing phase II carcinogen detoxifying enzymes; reduces oxidative stress and inflammation by decreasing reactive oxygen species (ROS) generation and downregulating cyclooxygenase (COX) and inflammatory cytokines; promotes cell proliferation arrest by modulating cell cycle regulatory machinery such as cyclins and cyclin-dependent kinases (CDKs); induces apoptosis of damaged or transformed cells by different mechanisms, including upregulation of the p53 tumor suppressor protein and BAX and downregulation of Bcl2 and survivin; suppresses angiogenesis, invasion and metastasis by inhibiting hypoxia-induced factor-1α (HIF-1α) and matrix metalloproteinases; targets hormone signaling due to its relevant antiestrogenic activity in hormone-dependent cancers; and reduces the risk of multidrug resistance (MDR) via multiple targets related to carcinogenesis and chemo/radioresistance [12,13]. Resveratrol has also been proposed as a pro-oxidant agent depending on the concentration, exposure time and cell type. The oxidative damage caused by this compound represents one of the cytotoxic mechanisms involved in tumor cell death [14,15].

Mutant p53 is associated with aggregation, which results in negative dominance and gain-of-function effects [16–19]. Novel compounds can directly target the interaction of p53 mutant aggregates with their p63 and p73 paralogues and with other transcription factors [17,20]. New compounds capable of intervening in the formation of aggregates can range from natural molecules such as resveratrol analogues, synthetic molecules such as Michael acceptors, small synthetic peptides, aptamers of nucleic acids and glycosaminoglycans [17]. These new drugs have great potential to represent radical innovation. In our recently published paper, we demonstrated that resveratrol

inhibits the aggregation of p53 mutants *in vitro*, in tumor cells and in xenotransplants implanted in nude mouse models (Figure 2) [21]. However, very high doses were required to exert the effect. We intend to use synthetic chemistry to produce resveratrol analogs with higher potency that can be used as pharmaceuticals.

**Figure 2.** Schematic representation of p53 inhibition by resveratrol. In (**1**), the resveratrol in vitro capacity of inhibition of both WT and mutant p53 aggregation is described. (**2**) When mutations in the TP53 gene appear, the protein produced is less stable and forms aggregates. These aggregates are related to a direct effect in cancer proliferation and migration that is inhibited by treatment with resveratrol (**3**). Otherwise, cancer progression occurs (**4**). Extracted from Ferraz da Costa, 2018 [21].

In another study, we described that transient transfection of the wild-type p53 gene causes H1299 cells (null to p53) to become more sensitive and responsive to the pro-apoptotic properties of resveratrol, similar to what was observed in MCF-7 cells. It was proposed that resveratrol could be used therapeutically in combination with other methods of promoting p53 activity in cells, such as gene therapy using the wild-type (WT) p53 gene or chemicals that restore p53 function [22].

In mammalian experimental models, resveratrol is extensively metabolized and quickly eliminated, resulting in poor bioavailability. After oral administration, resveratrol is absorbed by passive diffusion or by membrane transporters at the intestinal level and is then released in the bloodstream, where it can be detected as an unmodified or metabolized molecule [13,23]. Although 75% of resveratrol is absorbed through the oral route, only 1% is detected in the blood plasma after the whole metabolism [24]. In recent years, different methodological approaches and synthetic derivatives have been developed to improve resveratrol bioavailability. Many studies have been performed to synthesize new and more effective resveratrol analogs that display better pharmacokinetic properties, low toxicity and minimum side effects. The methoxylated, hydroxylated and halogenated resveratrol derivatives are more explored due to their beneficial biological activities and increased oral bioavailability [25]. Previous studies showed that methoxylation increases metabolic stability and the time length required for the molecules to reach the plasma concentration peak. Moreover, the substitution of hydroxyl groups of resveratrol to methoxyl groups substantially potentiated its therapeutic versatility. It was

also reported that the introduction of additional hydroxyl groups significantly increased the biological activity of resveratrol analogs [23,25]. In this review, we collect and present recent evidence in the literature regarding resveratrol derivatives and their anticancer effects, with an emphasis on the molecular mechanisms involved.

### *2.1. Methoxylated Resveratrol Derivatives*

### 2.1.1. Pterostilbene

Pterostilbene (*trans*-3,5-dimethoxy-40-hydroxystilbene) is a naturally occurring stilbene, found mainly in blueberries and grapes. It is a dimethylated derivative of resveratrol with comparable antioxidant, anti-inflammatory and anticarcinogenic properties [26]. Substituting its hydroxyl for a methoxyl group enhances the lipophilicity of pterostilbene, adding to its in vivo bioavailability and, thus, improving the biological activity of this compound compared to resveratrol [27]. A study designed to compare the bioavailability, pharmacokinetics and metabolism of resveratrol and pterostilbene following equimolar oral dosing administered in rats showed that resveratrol and pterostilbene were approximately 20% and 80% bioavailable, respectively [28]. Cumulative experimental data have noted that pterostilbene exerts multiple effects against a variety of cancer models through modulation of the cell cycle, induction of cell death, and inhibition of invasion and metastasis [26,29].

The first evidence of the anticancer properties of pterostilbene was demonstrated in a colon tumorigenesis model. Pterostilbene was shown to decrease the expression of inflammatory genes, such as iNOS in the colonic crypts and aberrant crypt foci (ACF) in rats, thus suggesting that its anti-inflammatory properties may be critical in colon cancer prevention [30]. Additionally, this compound inhibits preneoplastic lesions and adenomas in mouse colon by suppressing GSK3β phosphorylation and Wnt/β-catenin signaling pathway and reduces the expression of cyclin D1, vascular endothelial growth factor (VEGF) and matrix metalloproteinases (MMPs) [31]. It was also reported that pterostilbene showed significant dose-dependent antiproliferative and cytotoxic effects and inhibited Myc, beta-catenin and cyclin D gene expression in human colon cancer Caco-2 cells [32]. In gastric adenocarcinoma cells, pterostilbene inhibits cellular proliferation and leads cells to apoptosis by different pathways, such as caspase cascade activation and modulation of cell-cycle regulating proteins [33]. In a human model of hepatocellular carcinoma, pterostilbene suppresses tumor growth by interfering in the signal transduction pathways of NF-κB and on the expression of VEGF, matrix metalloproteinase-9 (MMP-9), AP-1 and mitogen-activated protein kinase (MAPK) [34,35]. Breast cancer stem cells isolated from MCF-7, which expresses the surface antigen CD44+/CD24–, were selectively eliminated by pterostilbene. Furthermore, this compound induces necrosis and inhibits mammosphere formation, increases the activity of paclitaxel, decreases CD44 expression, induces β-catenin phosphorylation through the inhibition of hedgehog/Akt/GSK3β signaling and decreases the expression of c-Myc and cyclin D1 [36]. More recent studies showed that pterostilbene is a promising agent against human papillomavirus (HPV) E6+ tumors tested in vitro and in vivo. In vitro, this compound downregulates the viral oncogene E6. On the other hand, in mouse TC1 tumors, in addition to inhibiting E6, pterostilbene suppressed VEGF and tumor development [37]. In acute lymphoblastic leukemia cell lines Jurkat and Molt-4, the potential of pterostilbene to modulate Fas, a member of the death-inducing family of tumor necrosis factor (TNF), was investigated. Pterostilbene increased both Fas mRNA and its cell surface levels, thus leading to apoptosis [38].

### 2.1.2. Trimethoxystilbene

Trimethoxystilbene (*trans*-3,4 ,5-trimethoxystilbene) is also a methylated resveratrol derivative, described as a potent chemopreventive agent, which promotes the induction of cell cycle arrest, reduces angiogenesis, inhibits cancer cell proliferation, increases apoptosis and decreases metastasis [25]. In MCF-7 breast cancer cells, this compound acts by inhibiting epithelial–mesenchymal transition (EMT), negatively modulatingβ-catenin nuclear translocation and phosphatidylinositol 3-kinase (PI3K)/protein

kinase B (AKT) signaling [39]. Its anticancer mechanisms on lung cancer cells involve apoptosis induction by activation of caspases 3 and 9 and poly (ADP-ribose) polymerase interruption [40]. Additionally, in human lung cancer (A549), trimethoxystilbene promotes a decrease in invasive, migratory and adhesive characteristics of these cells and modulates the mRNA levels that encode for MMP-2 protein [41]. When evaluating the effect of this resveratrol derivative in rat C6 and human T98G glioma cells, a massive accumulation of cells at the G2/M phase of the cell cycle and apoptosis via caspase-3 related to p53 tumor suppressor protein induction were observed [42]. Trimethoxystilbene is a more effective derivative than resveratrol in suppressing the growth of HepG2 hepatocellular carcinoma cells via induction of G2/M cell cycle arrest (by upregulation of cyclin B1) and apoptosis (by downregulation of Bcl-2) [43].

### 2.1.3. Tetramethoxystilbene

The modification of the resveratrol structure that generates its analogue tetramethoxystilbene improved its bioactivity by suppressing cell growth in prostate, colon, ovarian and hepatocellular cancer cells. This analogue demonstrates a higher activity on human melanoma A375 by decreasing cell proliferation after treatment using a lower dose (IC50 = 0.7 μM) than resveratrol (IC50 = 100 μM) [44]. Compared with the *trans* form of the 3,4,5,4 -tetramethoxystilbene resveratrol derivative compound, the *cis* form is ten times more potent at decreasing the growth of human WI38VA virally transformed fibroblasts [45]. Using a xenograft of human ovarian cancer (A2780 and SKOV-3) as a model to study the effect of *trans*-3,4,5,4 -tetramethoxystilbene, it was observed that treatment with this derivative is able to reduce tumor cell growth [46,47]. For breast cancer, it was demonstrated that proapoptotic proteins and voltage-dependent anion channel 1 (VDAC-1) expression were increased after treatment [48]. As a new approach to treat osteosarcoma cells resistant to paclitaxel and cisplatin, the use of tetramethoxystilbene decreases, in vitro and in vivo, the viability of resistant cells and induces massive apoptosis [49].

### 2.1.4. Pentamethoxystilbene

Pentamethoxystilbene is a hybrid molecule chemically synthesized and, similar to resveratrol, has low oral bioavailability but presents high intravenous bioavailability. In breast carcinoma cells (MCF-7), this derivative is a good antiproliferative candidate that acts through different pathways as a G1 cell cycle regulator, modulating cyclins E and D and retinoblastoma protein (pRb). It is a better suppressor agent for breast cancer cells than resveratrol or other methoxylated derivatives. At the IC50 concentration of this derivative (37.8 μM), treatment with resveratrol only reduces cell survival 20% [50]. In colon cancer, a better response was reached with this compound, with activation of apoptosis through cell cycle arrest in G2/M phase, polymerization of microtubules and finally caspase-induced apoptosis. Beyond apoptosis, the compound may decrease iNOS, β-catenin and cell proliferation [51,52]. No further studies have been published with this derivative since 2012.

### *2.2. Hydroxylated Resveratrol Derivatives*

### 2.2.1. Dihydroxystilbene

The resveratrol analogue 4,4 -dihydroxy-*trans*-stilbene (4,4 -DHS) was designed to improve resveratrol efficiency, both as an antioxidant and antiproliferative agent. 4,4 -DHS exhibits remarkably higher cytotoxicity than resveratrol against human promyelocytic leukemia (HL-60) cells [53]. 4,4 -DHS also inhibits the clonogenic efficiency of fibroblasts nine times more potently than resveratrol, although with a different mechanism. 4,4 -DHS predominantly induces an accumulation of cells in G1 phase, whereas resveratrol disturbs the G1/S phase transition. Furthermore, 4,4 -DHS increases p21 and p53 protein levels, whereas resveratrol leads to phosphorylation of the S-phase checkpoint protein Chk1 [54]. In a mouse lung cancer model, 2,3- and 4,4 -dihydroxystilbene (at 10 and 25 mg/kg, administered twice daily) inhibited tumor growth and metastasis. The antitumor and antimetastatic effects of these compounds were partly due to anti-lymphangiogenesis and the regulation of M2

macrophage activation and differentiation [55]. The cytotoxic action of 4,4 -DHS was also investigated in vitro in human neuroblastoma cell lines and in a mouse xenograft model of human neuroblastoma. The pharmacological action of 4,4 -DHS in the human neuroblastoma IMR32 cells was mediated by the destabilization of mitochondrial and lysosomal membranes, associated with modulation of several related pro- and anti-apoptotic cascades of proteins. Additionally, in the animal model, the oral administration of 4,4 -DHS for one month was well tolerated and demonstrated a greater therapeutic potential than resveratrol [56]. More recently, Saha et al. demonstrated that in melanoma cells, 4,4 -DHS acts by inducing apoptosis and cell cycle arrest in G1 phase and inhibiting cell proliferation. A significant reduction of melanoma tumors in a preclinical murine model was observed, and the antimetastatic effect of 4,4 -DHS was shown in a melanoma-mediated lung metastasis model in vivo [57]. In vivo assays performed in different mouse models of tumor xenografts demonstrated that 4,4 -DHS was able to disrupt the DNA replication pathway, leading to the apoptosis of pancreatic, ovarian and colorectal cancer cells [58].

### 2.2.2. Tetrahydroxystilbene

Tetrahydroxystilbene is a natural resveratrol analogue with multiple biological activities. In SK-Mel-28 melanoma cells, treatment with this compound induced apoptosis and inhibited cell proliferation [59,60]. In prostate cancer, its anticancer mechanisms involve JAK1 leading to STAT3 activation, which leads to a cytokine signal transduction pathway [61]. In liver and colon cancer cells, tetrahydroxystilbene arrests the cell cycle at G1 phase by modulating the cyclins pathway [62,63]. In human leukemia cells (U937), this compound induces massive apoptosis and leads to cell cycle arrest in G1 phase by regulating Bcl-2 and cIAP-2 (anti-apoptotic proteins). In cervix cancer, it acts by modulating p53 protein, thus leading cells to apoptosis [64,65]. In the last year, the 2,3,5,4 -tetrahydroxystilbene-2-O-β-D-glucoside (THSG) derivative was the major compound studied. In HT-29 colon adenocarcinoma cells, treatment with this compound reduced cell migration, invasion and adhesion, thus inhibiting metastasis. This is possible because THSG inhibits NF-κB pathway activation and consequently suppresses proteins involved in migration and invasion, such as MMP-2 and p-VE-cadherin, and ICAM-1 proteins involved in cell adhesion [66]. In vivo assays with THSG, using azoxymethane-induced colorectal cancer in rats, induced a 50% reduction in total colonic aberrant crypt foci by the inhibition of NF-κB pathway activation [67]. In MCF-7 breast cancer cells exposed to adriamycin and THSG, the vascular endothelial growth factor/phosphatidylinositol 3-kinase/Akt pathway was inhibited, triggering apoptosis by modulating Bcl-2 and caspase-3 [68].

### 2.2.3. Hexahydroxystilbene

Hexahydroxystilbene is a synthetic resveratrol derivative with higher biological activity [69]. When tested in breast cancer cells, hexahydroxystilbene induced apoptosis and inhibited cell proliferation by p53 protein accumulation and downregulation of mitochondrial superoxide dismutase [70]. In human colon cancer cells (HT-29), treatment with this compound leads to apoptosis and cell cycle arrest [71]. Similar results were observed in leukemia cells, with apoptosis induction by caspase activation pathways [72]. In vivo assays in a melanoma mouse model also demonstrate the induction of apoptosis pathway by upregulation of p21, downregulation of CDK-2 and cell cycle arrest at the G2/M phase [73]. Hexahydroxystilbene demonstrates an antiproliferative effect and accelerates senescence in cultured human peritoneal mesothelial cells by an oxidative stress-dependent mechanism. Treatment with 10 μM hexahydroxystilbene promoted an increase in 8-OHdG levels, a product of DNA oxidation, and a time-dependent increase in ROS release was also reported. On the other hand, soluble factors released by human peritoneal mesothelial cells that senesced prematurely in response to treatment promoted the growth of colorectal and pancreatic carcinomas in vitro [74]. No further studies have used this compound since 2013.

Although many studies have reported that stilbene compounds could play essential roles as chemotherapeutic agents by regulating multiple mechanisms and acting on different targets, further translational research is required to determine if the preclinical anticancer properties of these compounds, either alone or as part of combined therapies, are applicable in a clinical setting. The IC50 values for resveratrol and its derivatives vary in a wide range, as indicated in Table 1.


**Table 1.** IC50 values of resveratrol and derivatives in different study models.

### **3.** β**-Lapachone and Its Derivatives: The South American Promise for Cancer**

Quinones are widely distributed in nature, products of the secondary metabolism of several different species, with an incredible variety of biological responses [85,86]. Quinones may act as vitamins, antioxidants and are capable of stimulating antibacterial, antiallergic and anticancer effects, among others [87], which motivate their investigation as a therapeutic tool. Quinones can be cytotoxic through several mechanisms of action, including redox cycles, arylation of the thiol groups of proteins, intercalation, induction of breaks in the DNA chain, generation of free radicals and other ROS and bioreductive alkylation of critical cellular proteins and DNA via the formation of quinone methide. Besides basic research studies, quinone-based compounds are already used in the clinic. For instance, in cancer, doxorubicin, mitoxantrone and mitomycin C are used, among others [88,89]. Among all naphthoquinones described to date, three of them are widely known, mostly due to their anticancer effects and the story of their discovery—lapachol, β-lapachone and α-lapachone—of which lapachol (2-hydroxy-3-(3-methyl-2-butenyl-)-1,4-naphthoquinone, C15H14O3), with a molecular weight of 242.2738 g/mol, was isolated first from the heartwood of *Tabebuia impetiginosa*, a widespread tree species in Brazil and other South American countries [90]. No antitumor properties were reported for lapachol until Rao et al., in 1962, described a potent anticancer effect in rats [91]. Since then, a great interest in the research of the anticancer properties of lapachol and its derivatives, or structural isomers, α- and β-lapachone, has risen [92].

β-lapachone (3,4-dihydro-2,2-dimethyl-2H-naphthol[1,2-b]pyran-5,6-dione, C15H14O3), molecular weight 242.2738 g/mol, is an isomer of lapachol and has been described to promote several biological effects, such as anti-inflammatory, antibacterial and anti-*Trypanosoma* [93,94], and most important for this review, anticancer properties in different cancer types such as pancreatic cancer [95], breast cancer [96], hepatocellular carcinoma [97] and others. Several derivatives have been developed throughout the years, and it is noteworthy that a β-lapachone pro-drug, with commercial name ARQ-761, is in phase I/II of clinical studies for solid tumors [98].

### *3.1. Anticancer E*ff*ects*

As mentioned previously, there is a plethora of studies that demonstrate the ability of β-lapachone to induce cell death in several cancer cell lines, [95–97,99–101], but depending on the type of cancer, it is able to induce different types of cell death. Many studies demonstrate that β-lapachone is capable of inducing apoptosis [96,101,102] in cells such as HepG2, a hepatocellular carcinoma cell line [103], but, on the other hand, others demonstrate an ability to induce cell death via necroptosis, which is a type of organized necrosis [104–106]. As another example, Park et al., 2014 [97], showed that β-lapachone is capable of inducing this type of cell death in SK-Hep1, another hepatocellular carcinoma cell line.

Most anti-neoplastic drugs demonstrate a cytostatic effect, meaning that they are able to inhibit cell proliferation, and the ability of β-lapachone to prevent the proliferation of cancer cells has long been described [107]. IC50 values vary in a wide range, depending on the model tested (Table 2). As observed for the type of cell death that is induced by β-lapachone, the mode of cell cycle arrest is also dependent on the cell type under study. Dias et al. (2018) demonstrated that lapachone and its iodine derivatives induce cell cycle arrest in G2/M in human oral squamous cell carcinoma cells, and Lai et al. (1998) [108] showed cell cycle arrest in the S phase for a hepatoma cell line (HepA2).

There is also evidence of antitumoral effects of β-lapachone in preclinical studies. Wu et al. reported the promotion of heat shock protein 90 cleavage by β-lapachone, mediated by oxidative stress in NQO1-expressing cell lines. In the same work, in a mouse xenotransplant model, human lung cancer xenograft growth and angiogenesis were inhibited by β-lapachone treatment [109]. Kee et al. also demonstrated that β-lapachone is able to suppress lung metastasis of melanoma in an experimental mouse model [102].

**Table 2.** IC50 values of β-lapachone and derivatives in different study models.

\* IC50 values range for all derivatives shown in reference [112].

### *3.2. Mechanisms of Action*

### 3.2.1. ROS and NQO1

The primary mechanism of action of β-lapachone and its derivatives is the formation of ROS [92] through its processing by NAD(P)H quinone oxidoreductase 1 (NQO1). This enzyme is able to catalyze a futile redox cycle, leading to the formation of unstable hydroquinone, which is rapidly oxidized back to the original quinone under aerobic conditions [114]. The continuous redox cycles eventually oxidize a large number of reduced pyridine nucleotides, which form ROS [115]. This effect is quite robust, since one mol of β-lapachone is capable of generating 120 mol of superoxide in two min, consuming 60 mol of NAD(P)H [106], which results in a rapid depletion of intracellular NAD+ pool over 20 to 30 min [116]. This abnormal production of ROS leads to an increase in Ca++, depolarization of the mitochondrial membrane and a decrease in ATP synthesis. Therefore, in a general way, the activation of β-lapachone by NQO1 leads to cell death by apoptosis [117,118]. There are several studies that show that β-lapachone leads to the formation of ROS in cancer cells, such as Park et al., in 2014, who report that the increase of ROS is capable of inducing cell death of hepatocellular carcinoma cells (SK-Hep1) [97]. In 2011, the same group showed that ROS were involved in β-lapachone-induced autophagy in glioma cells (U87 MG). Bey et al. (2013) showed that H2O2 is the primary obligate ROS species necessary for β-lapachone breast cancer cell cytotoxicity through lipid peroxidation, which damages cellular membranes and organelles [106].

It is important to note that several types of solid tumors, such as cholangiocarcinoma [119], lung [120], pancreas [121], breast [122] and squamous cell carcinoma of the uterine cervix [123], have high expression of NQO1, and there are studies that demonstrate a β-lapachone preferential tropism for NQO1-positive cells [124].

Liver tumors are a very interesting case, since normal hepatocytes do not express NQO1 [125,126], but preneoplastic lesions and hepatic tumors demonstrate the presence of this enzyme [127,128]. Thus, this differentiated expression may be very important in the development of targeted therapies since it induces cell death of neoplastic or preneoplastic cells, greatly reducing side effects on healthy liver cells. Additionally, hyperthermia has been reported as an enhancer of β-lapachone effects due to the increase in NQO1 levels after heat shock and its stabilization by HSP70 [129–132]. Finally, it is noteworthy that NQO1-positive breast cancer cells correlate with the malignancy of the disease and could be used as a prognostic biomarker for breast cancer [124].

### 3.2.2. Topoisomerase Inhibition

One of the first mechanisms of action reported for β-lapachone was its role as a topoisomerase I modulator. Earlier thought of as a topoisomerase activator [133–135], it was later shown to act as an inhibitor, through the demonstration that β-lapachone inhibits the catalytic activity of topoisomerase I (purified from calf thymus and human cells), through its direct binding, since the incubation of topoisomerase I with β-lapachone (before adding DNA) considerably increased its inhibition, but the incubation of topoisomerase I with DNA prior to the treatment did not show any effect [136]. Other studies refer to similar effects in cell lines of different cancers, such as prostate cancer, breast cancer and leukemia [96,137,138].

### 3.2.3. p53

A number of studies have demonstrated β-lapachone effects on p53 independently of the cell p53 status (no expression or expression of wild-type or mutant p53); in all cases, cells can be sensitive to the effects of β-lapachone [107,139,140]. Huang Pardee, in 1999 [107], reported that β-lapachone had the ability to drastically reduce levels of mutant p53 in colon cancer cells, although it did not alter the expression levels of wild-type p53. In addition, p53 can also be regulated by β-lapachone through its phosphorylation and consequent activation, with no modification in its expression levels [141]. Finally, Pink et al. (2000) reported the β-lapachone-mediated activation of a cysteine-protease capable of digesting several cellular proteins, including p53 [142].

### 3.2.4. Other Cellular or Molecular Pathways

Yu et al. demonstrated the anti-tumor effect of β-lapachone on breast cancer tumors with a variation in the phosphorylation of AKT, 4EBP-1 and S6, which are related to the mTOR pathway and is also related to the apoptotic activity of this compound in gastric carcinoma cells [101]. Additionally, Wu et al., in 2016 [109], demonstrated that the effect of β-lapachone on reducing the growth of lung cancer cell tumors is related to AKT. In addition to the mTOR pathway, E-cadherin was altered in

tumors, and Kim et al. (2007) [102] suggested that β-lapachone inhibits the progression and metastasis of hepatocellular carcinoma by increasing the expression of this protein and other proteins of the mTOR pathway. Additionally, the inactivation of the Akt/mTOR pathway was again attributed to β-lapachone, promoting the inhibition of EMT transition in NQO1-positive cells.

### *3.3. Strategies to Overcome* β*-Lapachone Bioavailability and Toxicity Issues: Drug Delivery and Derivatives Synthesis*

Even though β-lapachone is a promising anticancer drug, its low bioavailability represents a limitation for clinical use due to low solubility in water and gastrointestinal fluids [143]. There is also a concern with the low concentration reached in the target cells and systemic toxicity, since β-lapachone displays a general distribution pattern and a dose limitation because of the risk of methemoglobinemia through the generation of nonspecific ROS at high doses [116,143].

As mentioned before, β-lapachone (Table 2) has a molecular weight of 242.29 g/mol; it is a small molecule, nonionized in the intestinal system, with pH-independent solubility. The experimental solubility value is very similar to the theoretical value of 48.33 μg/mL [112,144]. β-lapachone has the potential to be orally administered; its estimated oral fraction absorbed and intestinal effective permeability value is 85% [145]; however, a bioavailability of 15.5% through oral administration was shown in rats, probably due to broad metabolic first-pass degradation in the liver, intestines and low solubility in water, with a slow dissolution rate in the intestinal tissue [144]. Moreover, preclinical studies demonstrate that, to promote a better absorption of β-lapachone, different formulations would be necessary, and due to these findings, along with its first-pass metabolism, it was considered a difficult drug for oral administration [144,146].

Due to the low availability and unspecific toxicity of β-lapachone, there is a constant need for the development of new drug delivery systems to increase bioavailability to promote its use. There are two classical systems of β-lapachone delivery, cyclodextrin inclusion complexes and liposomes, although most studies are concerned with physical–chemical and absorption properties without any cancer therapy approach. A thorough review on this subject was written by Ferreira et al., 2016 [147]. Here, we focus on anticancer approaches.

The intricate structure of cyclodextrin allows the formation of β-lapachone/cyclodextrin inclusion complexes because of its hydrophobic core [145,148]. These inclusion complexes are capable of increasing drug bioavailability, altering their permeability, dissolution properties or both [149]. Seoane et al. (2013) used a methylated-β-cyclodextrin/poloxamer 407 mixture to create a delivery system of β-lapachone and evaluated its antitumoral activity. This system has the particularity of forming a gel above 29 ◦C, which facilitates intratumoral and extended drug delivery. MCF-7 tumor-bearing mice treated with this system, by intratumoral injection, showed a reduction of tumor volume without apparent liver and kidney toxicity [150].

Liposomes (one or multiple layers of phospholipids) are very interesting delivery systems because they can be used for both hydrophobic and hydrophilic drugs and display remote drug loading, homogeneous particle size, long-circulating stability, specific release and the ability to lower drug toxicity [151]. Liposomes with β-lapachone, in different mixed micellar formulations of phosphatidylcholine, sodium deoxycholate and sodium lauryl sulfate (SLS), showed the ability to increase gastrointestinal absorption at different sites (particularly in the large intestines) due to β-lapachone solubilization and interaction with the intestinal membrane [152].

Recently, there has been an increased exploration of nanosystems or nanoparticle delivery systems to surpass the limitations of these delivery systems and increase specificity and cytotoxicity to cancer cells. Dai et al. (2019) [153] developed a nanosystem with a charge-reversal ability and self-amplifiable drug release system that encapsulated β-lapachone in a pH/ROS cascade-responsive polymeric prodrug micelle. They showed that this system is capable of effectively increasing cell uptake and specific delivery through acidity-activating charge conversion and ROS-response drug release. Upon the uptake of β-lapachone, ROS formation was increased (NQO1-mediated), which lowered the cell ATP levels and consequently reduced P-gp-mediated drug efflux, decreasing multidrug resistance (MDR). This caused a massive reduction of MCF-7 tumors treated with this β-lapachone delivery system and low systemic toxicity [153].

Another strategy to increase the therapeutic efficacy of β-lapachone, through the reduction of MDR, is to promote its codelivery with another drug, such as doxorubicin. This was an option used by Li et al., demonstrating that a nanostructured lipid carrier (NLC) codelivering β-lapachone and doxorubicin had a higher therapeutic efficacy in breast cancer tumor-bearing mice, leading the authors to propose this as a possible strategy to overcome MDR in breast cancer [154].

A nanoparticle developed by Yin et al., a ROS-responsive block copolymer prodrug that self-assembles into polymeric micelles that encapsulate β-lapachone, increased tumor-specific ROS formation after intravenous administration in tumor-bearing mice. This increased ROS level and triggered drug release, allowing maximization of therapeutic efficacy, suppression of tumor growth and minimization of systemic toxicity [155].

There are three main strategies to develop β-lapachone derivatives or analogues—A- and C-ring modifications and redox center modifications—which can be obtained by copper-catalyzed azide-alkalyne cycloaddition, palladium-catalyzed cross couplings and heterocyclization reactions [156]. The great progress seen in this field is due to essential contributions from Brazilian research groups, such as Vitor Francisco Ferreira and Eufrânio Nunes da Silva Júnior [156–161]. However, most studies of β-lapachone derivatives or analogues are composed of a series of compounds with different modifications and analysis of the possible anticancer effect in cancer cell lines [89,156,161]. Nevertheless, there are more in-depth studies that attempt to understand the mechanisms of action of the compounds and eventually test them on experimental tumor models.

Recently, Dias et al. (2018) [113] showed that β-lapachone and its 3-iodine derivatives (3-I-α-lapachone and 3-I-β-lapachone) were able to induce significant cytotoxicity against different types of cancer cells (Table 2), with cell cycle arrest in G2/M, DNA fragmentation, increase in apoptosis protein levels and morphology, and production of ROS. This work also demonstrated that these compounds were able to reduce tumor burden for mice xenotransplanted with breast adenocarcinoma cells, without any alteration of biochemical, hematological or histological parameters of the treated mice, showing a nonsignificant systemic toxicity. Li et al. described a new class of β-lapachone derivatives, naphtho[2,1-d]oxazole-4,5-diones (Table 2), with higher solubility and comparable activities, both in vitro and in vivo (xenotransplants), against NQO1-positive A549 lung cancer cells [112]. The compounds β-lap-dC3 and -dC6 are prodrug diester derivatives of β-lapachone. When encapsulated in PEG-β-PLA micelles, they were described to be more efficient in loading micelles than β-lapachone itself, and β-lap-dC3 improved the survival rate of NSCLC xenograft-bearing mice, increasing β-lapachone concentration in the target tissues, which makes it a promising therapy to be developed against NQO1-positive cells [162].

Finally, the sensitization of cells promoted by different energy sources, such as light or radiation, in combination with β-lapachone and its derivatives has also been described. Lamberti et al. (2018) showed the synergism between halogenated derivatives of β-lapachone and photodynamic therapy in melanoma cells with positive results due to the upregulation of NQO1 expression [163]. Ionizing radiation at low doses was applied in combination with sublethal doses of β-lapachone in non-small-cell lung cancer (NSCLC) cell lines and in xenograft models in vivo [164]. Additionally, the massive release of ROS promoted by ionizing radiation was applied together with nontoxic β-lapachone doses to head and neck patient samples [165]. These results suggested the use of this combination to increase the efficacy of radiotherapy in NQO1-positive tumors and shall be tested in clinical trials in the near future.

### **4. Conclusions**

Both of the molecules reviewed here, β-lapachone and resveratrol, are paradoxical. While they are naturally occurring products, or at least derived from natural sources, they can be radically improved by human manipulation. They represent a case study for natural products that can be enhanced by structural manipulation and/or delivery systems. In this paper, we intended to compile the available information about β-lapachone and resveratrol, with a special focus on what has been found in recent years and on the possible impact in cancer therapy, also showing how a bioactive compound/molecule can be found in natural resources, which are a possible reservoir of new therapies. However, it is essential not to be dismissive of the importance of manipulating natural molecules to increase their efficiency and specificity. The results obtained with β-lapachone, resveratrol and their analogues for cancer therapy are promising, especially if we manage to improve their specificity for cancer cells, with less systemic toxicity (a major problem for most chemotherapies) and their delivery, particularly by making them orally bioavailable for patient comfort. The similarities and differences between the two groups of compounds can be observed through the comparison between the values shown in Tables 1 and 2 and the properties listed in the supplementary material tables (Tables S1–S4).

**Supplementary Materials:** The following are available online, Table S1: Bioactivity prediction (druglikeness\*) of β-lapachone and its derivatives for specific biological activities; Table S2: Bioactivity prediction (druglikeness\*) of resveratrol and its derivatives for specific biological activities; Table S3: Chemical and pharmacological properties of Lapachol, β-lapachone and derivatives; Table S4: Chemical and pharmacological properties of resveratrol and derivatives.

**Author Contributions:** D.C.F.d.C., L.P.R., V.F.F. and J.L.S. wrote and edited the text. M.M.D.d.C.M.-D. and G.D.d.S.F. wrote the text. All authors have read and agreed to the published version of the manuscript.

**Funding:** We thank Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ), Fundação do Câncer and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for the support.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Review* **Molecular Mechanisms of the Anti-Cancer E**ff**ects of Isothiocyanates from Cruciferous Vegetables in Bladder Cancer**

### **Tomhiro Mastuo, Yasuyoshi Miyata \*, Tsutomu Yuno, Yuta Mukae, Asato Otsubo, Kensuke Mitsunari, Kojiro Ohba and Hideki Sakai**

Department of Urology, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki 852-8501, Japan; tomozo1228@hotmail.com (T.M.); t.yuno@nagasaki-u.ac.jp (T.Y.); ytmk\_n2@yahoo.co.jp (Y.M.); a.06131dpsc@gmail.com (A.O.); ken.mitsunari@gmail.com (K.M.); ohba-k@nagasaki-u.ac.jp (K.O.); hsakai@nagasaki-u.ac.jp (H.S.)

**\*** Correspondence: yasu-myt@nagasaki-u.ac.jp; Tel.: +81 95 819 7340; Fax: +81 95 819 7343

Received: 10 January 2020; Accepted: 28 January 2020; Published: 29 January 2020

**Abstract:** Bladder cancer (BC) is a representative of urological cancer with a high recurrence and metastasis potential. Currently, cisplatin-based chemotherapy and immune checkpoint inhibitors are used as standard therapy in patients with advanced/metastatic BC. However, these therapies often show severe adverse events, and prolongation of survival is unsatisfactory. Therefore, a treatment strategy using natural compounds is of great interest. In this review, we focused on the anti-cancer effects of isothiocyanates (ITCs) derived from cruciferous vegetables, which are widely cultivated and consumed in many regions worldwide. Specifically, we discuss the anti-cancer effects of four ITC compounds—allyl isothiocyanate, benzyl isothiocyanate, sulforaphane, and phenethyl isothiocyanate—in BC; the molecular mechanisms underlying their anti-cancer effects; current trends and future direction of ITC-based treatment strategies; and the carcinogenic potential of ITCs. We also discuss the advantages and limitations of each ITC in BC treatment, furthering the consideration of ITCs in treatment strategies and for improving the prognosis of patients with BC.

**Keywords:** allyl isothiocyanate; benzyl isothiocyanate; sulforaphane; phenethyl isothiocyanate; bladder cancer

### **1. Introduction**

Bladder cancer (BC) is recognized as a representative of urological cancer (UC), but it has specific pathological characteristics and treatment strategies. BC shows a high recurrence and metastasis potential, even if radical operation is performed [1,2]. Regarding treatment strategies, only two regimens have been approved as effective methods: platinum-based chemotherapy, including gemcitabine and cisplatin combination therapy and MVAC therapy, and immune checkpoint inhibitors, such as pembrolizumab [3–5]. Unfortunately, these regimens have shown relatively severe adverse events, including renal dysfunction, neutropenia, and immunogenic abnormalities, and it is currently difficult to predict therapeutic effects [3,4,6]. Although there is an improved prognosis of BC patients undergoing these therapies, prolongation of survival is far from satisfactory, especially in patients with advanced/metastatic disease. Currently, new anti-cancer drugs for UC are under development, some of which are expected to be approved for the treatment of advanced/metastatic BC in the near feature [7,8]. However, there is little information on the safety and adverse events of these new drugs.

Based on these facts, as well as the artificially produced anti-cancer drugs, treatment strategies employing natural compounds have garnered special interest for various types of malignancies. There is a general consensus that considering treatment strategies using natural product(s) is important

for improving the quality of life and prognosis of cancer patients. In fact, there is a report that curcumin, a natural occurring polyphenol derived from turmeric (*Curcuma longa*), had anti-cancer effects via suppression of cancer cell proliferation, invasion, and metastasis in lung cancer cells [9]. The authors also showed that regulation of microRNA expression plays important roles in the curcumin-induced anti-cancer effects [9]. Furthermore, Yiqi Huayu Jiedu decoction, which comprises various Chinese herbs and natural compounds, was reported to increase the anti-cancer effects of standard chemotherapy in patients with stage III gastric cancer after radical gastrectomy and improve the quality of life of patients [10]. In addition to these reports, several reviews and *in vivo* and *in vitro* studies have demonstrated that a variety of natural products possess anti-cancer potential and can avoid the undesirable effects of standard therapies of malignancies [11–14].

Regarding BC, multiple studies showed that a variety of natural products have anti-cancer effects and maintain quality of life, including *Evodia rutaecarpa* or curcumin [15,16]. Several reports also demonstrated the relationship between natural products and prevention of carcinogenesis, clinicopathological features, and anti-cancer effects in BC [17,18]. In previous studies, we showed the anti-cancer effects, clinical usefulness, and pathological mechanisms of green tea polyphenol or royal jelly in urological cancers, including BC [19–25]. Although green tea and royal jelly are eaten in some parts of Asia and Western countries, this is not the case globally. In contrast, cruciferous vegetables, such as broccoli, kale, cauliflower, bock choy, and horseradish, are widely cultivated in many regions and are commonly eaten worldwide. Isothiocyanates (ITCs) are naturally occurring products of cruciferous vegetables, and researchers have investigated their health benefits and efficacy in the treatment of various diseases, including malignancies. Although some reviews mentioned the anti-cancer effects of ITCs in BC, there is relatively little comprehensive information on the molecular mechanisms of the ITC anti-cancer effects in BC cells [26–28]. In addition, comprehensive information on each ITC member, including allyl isothiocyanate (AITC), benzyl isothiocyanate (BITC), sulforaphane (SFN), and phenethyl isothiocyanate (PEITC), is limited. Therefore, in this review, we discuss the anti-cancer effects and efficacy of ITCs in BC cells obtained by *in vivo* and *in vitro* studies. In particular, we focus on the changes in malignant behaviors and cancer-related molecules by ITC members in BC cells. Furthermore, we provide future direction of ITC-based therapy for patients with BC.

### **2. Isothiocyanates in Cruciferous Vegetables**

Cruciferous vegetables are classified in the family Brassicaceae/Cruciferae. Several *in vivo*, *in vitro*, and epidemiological studies have shown that cruciferous vegetables inhibit carcinogenesis of BC [27,29,30]. However, in contrast to these reports, a study suggested that cruciferous vegetable intake is not significantly associated with reduced BC risk [31]. Thus, there are controversial results regarding the relationship between cruciferous vegetables and cancer risk in UC. Additionally, there is limited knowledge on the molecular mechanisms of the anti-cancer effects of cruciferous vegetables.

ITCs are naturally present in cruciferous vegetables and are produced by the hydrolysis of glucosinolates [32]. There is a general agreement that ITCs are beneficial for human health via various mechanisms, such as their anti-microbial activity, prevention of cardiovascular disease, and improvement of fasting glucose levels [33,34]. ITCs are also reported to exhibit anti-carcinogenic activities in various cancer types, including BC [27,33,35,36]. ITCs include the compounds AITC, BITC, SFN, and PEITC (Figure 1), each of which has multiple activities, including anti-cancer effects [37]. In the following sections, we will present the changes in malignant aggressiveness and molecular expression/activity in BC by each ITC member.

**Figure 1.** Structures of the isothiocyanate members.

### *2.1. Allyl Isothiocyanate*

AITC is a volatile and water-insoluble compound derived from various cruciferous vegetables that exhibits multiple functions, such as anti-inflammation, neuroprotection, and anti-bacterial activity [38,39]. In addition, AITC is reported to have anti-cancer effects in various types of malignancies [40–42]. However, a study showed that AITC had no significant inhibitory effects on cell proliferation or stimulation of apoptosis in the human breast cancer cell line MDA-MB-231 [43]. Nevertheless, other studies have highlighted the potential advantages of AITC in BC treatment, as the major route of excretion of orally administered AITC is through urine, demonstrating relatively high bioavailability in urine and bladder tissue compared with other organs [44,45]. In this section, we will introduce the anti-cancer effects of AITC in BC and provide future direction for novel treatment strategies using AITC.

### 2.1.1. In Vitro Studies

The anti-cancer effects of AITC and its molecular mechanisms have been investigated in various BC cell lines; for instance, AITC was found to lead to morphological changes and inhibit the cell proliferation of the human BC cell lines RT4 and T24 [46]. AITC was also reported to affect cell cycle arrest and apoptosis of RT4 and T24 cells [44]. Moreover, the cytotoxic effects of AITC were confirmed in another BC cell line (UM-UC-3 cells) [47], and the percentages of apoptotic cells increased in an AITC dose-dependent manner in three different BC cell lines (UM-UC-3, UM-UC-6, and T24) [48]. This same study demonstrated that AITC-induced apoptosis is mediated by a mitochondrion-mediated system, including activation of caspase-9, caspase-3, lamin B1, and poly ADP-ribose polymerase (PARP) as well as Bcl-2 phosphorylation at Ser-70 by c-Jun N-terminal kinase (JNK) [48].

These anti-cancer effects, such as anti-proliferation and pro-apoptosis, of AITC are speculated to be independent from TP53—an important regulator of cell death—because RT4 cells possess wild-type TP53, while T24 and UM-UC-3 cells possess mutated TP53 [49]. Moreover, the anti-cancer effects of AITC involved different molecular mechanisms between RT4 and T24 cells. In RT4 cells, AITC treatment increased S100P and Bax levels and decreased Bcl-2 levels; meanwhile, Bax, Bcl-2, and anillin levels increased while S100P levels decreased in T24 cells [46]. The Bax/Bcl-2 pathway is speculated to be a key modulator of AITC in RT4 cells, with anillin and S100P mainly functioning in this system [46]. Thus, these *in vitro* studies demonstrated that the anti-cancer effects of AITC in BC cells are dependent on the pathological and molecular characteristics of cancer cells.

Molecular mechanisms of the AITC-induced anti-cancer effect in BC cells are shown in Table 1. To our knowledge, there are only two *in vitro* studies on this topic, warranting further research to discuss treatment strategies using ITCs.


**Table 1.** *In vitro* molecular mechanism of the anti-cancer effects of allyl isothiocyanate.

Bcl, B-cell lymphoma-2; Bax, Bcl-2-associated X protein.

### 2.1.2. *N*-acetylcysteine Conjugate Allyl Isothiocyanate

AITC is mainly excreted in urine as *N*-acetylcysteine conjugate (NAC-AITC). In human and rat BC cells (UM-UC-3 and AY-27 cells, respectively), NAC-AITC inhibits cell proliferation and regulates cell cycle arrest and apoptosis [50]. The anti-cancer effects of NAC-AITC were found associated with downregulation of α-tubulin, β-tubulin/ and vascular endothelial growth factor and activation of caspase-3. Moreover, the authors conclude that the anti-cancer effects of NAC-AITC, including prevention and treatment of cancer are superior to AITC in terms of pharmacokinetic and physical properties. Similar anti-tumor growth activity was found in an orthotopic rat BC model, wherein bladder tumor weight in the NAC-AITC group is significantly lower than that in control (p = 0.0213) [50]. In addition, NAC-AITC suppressed muscle invasion of BC cells (NAC-AITC group = 30%; control = 79%). Similar to the BC cell lines, α- and β-tubulin, vascular endothelial growth factor, and cleaved caspase-3 were found associated with the *in vivo* anti-cancer effects [50].

### 2.1.3. In Vivo Studies

Dietary administration of a freeze-dried, aqueous extract of broccoli sprouts that included AITC was found to reduce the incidence, multiplicity, and size of BC in an *N*-butyl-*N*-(4-hydroxybutyl) nitrosamine (BBN)-induced rat BC model [51]. However, the detailed molecular mechanisms of this anti-cancer effect were not clearly defined. Another study showed that oral intake of AITC-rich mustard seed powder inhibits tumor growth and muscle invasion in an orthotopic rat BC model via regulation of apoptosis, cell cycle, and angiogenic potential [52]; downregulation of vascular endothelial growth factor and cyclin B1 and upregulation of caspase-3 and cleavage of PARP were found associated with these anti-cancer effects [52].

When AITC is stably stored as its glucosinolate precursor (sinigrin) in mustard seed powder (MSP-1), a study revealed that sinigrin itself is not bioactive, whereas hydrated MSP-1 leads to apoptosis and G2/M phase arrest in bladder cancer cell lines *in vitro*. In an orthotopic rat bladder cancer model, oral MSP-1 inhibited bladder cancer growth by 34.5% (P < 0.05) and blocked muscle invasion by 100%. The anti-cancer activity of AITC delivered as MSP-1 appears to be more robust than that of pure AITC. Therefore, MSP-1 may be an attractive delivery vehicle for AITC, as it strongly inhibits bladder cancer development and progression [52].

2.1.4. Combination Therapy of Allyl Isothiocyanate and Conventional Anti-cancer Agents

The cyclooxygenase (COX)-2-plastaglandin (PG) E2-system is an important pathological mechanism of carcinogenesis, tumor growth, and progression in UC [53–55]. Therefore, COX-2 inhibitors have been suggested as chemoprotective and therapeutic agents in a variety of cancers [56–58]. The synergistic effects of a combination of COX-2 inhibitors and other standard therapy have also been reported [59,60]. Celecoxib, a selective COX-2 inhibitor, is used for various pathological conditions worldwide. Thus, to clarify the anti-cancer effects of a combination of celecoxib and AITC, *in vitro* studies employing AY-27 bladder cancer cells and *in vivo* studies with the F344/AY-27 rat bladder urothelial cell carcinoma model were performed [61]. *In vitro*, AITC first showed no significant impact on COX-2 expression, and PGE2 production was confirmed. However, when the growth inhibitory effects of AITC and celecoxib were analyzed, growth inhibition of AY-27 cells by AITC was not altered by celecoxib addition. The authors thus speculated that the COX-2-mediated anti-tumor growth effects of celecoxib did not reach detectable levels due to excessive dilution of PGE2 in the culture medium. On the other hand, *in vivo* studies employing an animal model with orthotopic BC showed that combination therapy of AITC (1 mg/Kg) and celecoxib (10 mg/Kg) suppresses tumor growth and muscle invasion and that these anti-cancer effects are stronger compared with those of AITC or celecoxib alone. Inhibition of tumor-related angiogenesis regulated by vascular endothelial growth was found to play a crucial role in these anti-cancer effects.

Another combination therapy employed AITC and cisplatin, a standard anti-cancer drug for patients with BC [62]. *In vitro* studies with lung cancer cells (HOP62) and ovarian cancer cells (2008) showed that the variabilities of both cancer cell lines are significantly inhibited by a combination of AITC and cisplatin, whose inhibitory effects are stronger compared with those of AITC or cisplatin alone. The anti-proliferative effects were confirmed by colony formation assays, and when relationships between cell death and the combination therapy were examined, levels of pro-apoptotic molecules (caspase-3) were found increased and anti-apoptotic molecules (Bcl-2 and survivin) decreased; thus, this combination can suppress tumor growth *in vitro*. Mechanistically, regulation of cell cycle, β-tubulin depletion, and microtubule dysfunction are associated with the anti-cancer effects of AITC and cisplatin. Finally, the combination index of ATIC and cisplatin in lung cancer cells indicates a synergistic interaction. Indeed, *in vivo* studies with A549-derived lung cancer xenograft tumor models showed decreases in tumor volumes after combination therapy (AITC = 50 mg/Kg and cisplatin = 6 mg/kg), whereas tumor volumes increased after AITC (50 mg/Kg) or cisplatin (6 mg/kg) monotherapy. Overall, the anti-cancer effect parameters (i.e., maximum tumor growth inhibition, tumor doubling time, and frequency of partial response and complete response) are remarkably better after combination therapy than after either monotherapy. Furthermore, AITC + cisplatin therapy exhibits no toxicity, including maximum weight loss of pretreatment bodyweight.

### 2.1.5. Clinical Trials and Future Direction of Allyl Isothiocyanate-Based Therapy

Recently, an *in vitro* study employing the macrophage cell line RAW 264.7 and human BC cell line HT1376 was conducted to clarify the anti-inflammatory activity and anti-cancer effect of AITC nanoparticles [63]. The results showed that AITC nanoparticles inhibit cancer cell proliferation and migration; however, these anti-cancer effects are dependent on AITC concentration; inhibition of cancer cell proliferation and migration is achieved at 70 mg L−<sup>1</sup> and 8.75 mg L−<sup>1</sup> of AITC nanoparticles, respectively. AITC nanoparticles were also found to suppress production of lipopolysaccharide-induced tumor necrosis factor (TNF)-α, interleukin (IL)-6, nitric oxide (NO), and inducible NO synthase in macrophage cells, and their anti-inflammatory effects are stronger than those of AITC or nanoparticles alone. The authors thus suggested that AITC nanoparticles can be a valuable treatment strategy for BC via their regulation of inflammation, immunity, and oxidative stress.

Other novel strategies employing AITC are currently under development. For example, the anti-cancer effects of AITC-conjugated silicon quantum dots were examined in human umbilical vein endothelial cells (HUVECs) and human hepatocellular carcinoma cells (HepG2) [64]. Interestingly, high doses of AITC (40–320 μM) were found to significantly inhibit HepG2 cell viability, whereas low doses (5 μM) significantly stimulated cancer cell viability. Similar trends were confirmed for cancer cell migration (inhibition at 20 μM AITC and stimulation at 2.5 μM) and angiogenesis (HUVEC

tube formation ability is suppressed at > 5 μM but stimulated at even lower doses of 1.25 and 2.5 μM AITC). Thus, there is a possibility that the anti-cancer effects of AITC are dependent on its concentration and that low concentrations of AITC may have detrimental effects via increased cancer cell proliferation, migration, and angiogenesis in hepatocellular carcinoma. The authors further showed that AITC-conjugated silicon quantum dots overcame the limitation of AITC in the same analysis. Therefore, AITC-conjugated silicon quantum dots are suggested as a useful drug delivery system for AITC in cancer patients. Although there are no data in BC, AITC is also predicted to have biphasic effects of anti-cancer effects and angiogenesis. Therefore, we suggest additional *in vivo* and *in vitro* studies of the silicon quantum dot system in BC to elucidate new treatment strategies for patients.

### *2.2. Benzyl Isothiocyanate*

Similar to other ITC members, BITC has immunomodulatory, anti-microbial, and anti-oxidative activities under various pathological conditions [34,65,66] Several studies have also shown that BITC possesses anti-cancer and chemopreventive effects in various types of malignancies [67–69]. However, there is limited information on its anti-cancer effects and molecular mechanisms in UC.

### 2.2.1. In Vitro Studies

Similar to AITC, BITC has shown anti-proliferative and pro-apoptotic activity in BC cells [70,71]; however, the pro-apoptotic activity of BITC is stronger compared with that of other ITC members, including AITC and SFN. Moreover, caspase-9 is the main regulator of BITC-induced apoptosis in UM-UC-3 cells, although all ITC members exhibit pro-apoptotic activities via activation of caspase-3, 8, and 9 [70,72]. Additionally, mitochondrial activities are targets of BITC, and BITC-induced changes are regulated by various members of the Bcl-2 family, including Bcl-2, Bax, Bak, and Bcl-xl [70].

As previously mentioned, ITCs are primarily disposed and concentrated in the urine as NAC conjugates. UC originates from urothelial cells and is constantly exposed to urine in the urinary tract. Therefore, studies have focused on the anti-cancer effects of NAC-conjugated BITC in BC cells [71] and found that it suppresses BC cell growth through anti-proliferative and pro-apoptotic activities. Activation of caspase-3, 8, and 9; cell cycle arrest in phases S and G2/M; and regulation of Cdc25C were associated with the anti-proliferative function of NAC-conjugated BITC. The authors confirmed, however, that longer treatment durations or higher doses of NAC-conjugated BITC are necessary to exert similar effects as those of BITC.

miRNAs are major modulators of carcinogenesis, malignant aggressiveness, and outcome in UC [73,74] and several miRNAs are closely associated with cisplatin sensitivity of BC cells [75]. miR-99a-5p, a tumor suppressor, exhibits anti-proliferative and pro-apoptotic activities in UC [76–78]. One study demonstrated that BITC treatment upregulates miR-99a-5p expression in the BC cell lines 5637 and T24 [76], which leads to decreased mRNA and protein levels of IGF-1R, FGF-R3, and mTOR in both BC cell lines. The authors also demonstrated a molecular mechanism associated with regulation of BC cell survival and apoptosis by BITC. Taken together, these findings indicate that BITC exhibits anti-cancer effects via regulation of cell survival in UC. Another study elucidated the anti-cancer effects of BITC-induced miR-99a expression in BC cells [79] and reported that BITC enhances miR-99a expression in 5637 and T24 BC cells, which is associated with ERK activation and nuclear transcriptional activation of c-Jun/(activator protein) AP-1. Thus, the authors suggested that BITC stimulates miR-99a expression via regulation of the ERK/AP-1 pathway in BC and demonstrated the anti-cancer effects of miR-99a in UC. Nevertheless, there is a general consensus that the anti-carcinogenic and anti-cancer effects of miR-99a represent a complex mechanism in BC cells [80–82]. Therefore, this information is useful for understanding the biological function of BITC in BC. The molecular mechanisms of the anti-cancer effects of BITC are shown in Table 2.


**Table 2.** *In vitro* molecular mechanisms of the anti-cancer effects of benzyl isothiocyanate.

IGF1R, insulin-like growth factor 1 receptor; FGFR, fibroblast growth factor receptor; mTOR, mechanistic target of rapamycin; Bcl, B-cell lymphoma-2; Bax, Bcl-2-associated X protein; Bak, BCL2-antagonist/killer.

### 2.2.2. In Vivo Studies

In a rat model of BBN-induced BC, oral intake of BITC suppressed the incidence of neoplastic pathological changes, such as dysplasia, papilloma, and carcinoma, and multiplicities in a dose-dependent manner (10, 100, or 1,000 ppm BITC) [83]. Notably, epithelial hyperplasia of the bladder was found in rats treated with 100 or 1,000 ppm BITC without BBN [83]. The same researchers also demonstrated the carcinogenic potential of BITC in this BC animal model [84]. Therefore, the toxicity and risk of BITC in BC treatment should be considered; this is further detailed later in the text (see Section 2.1).

### 2.2.3. Combination Therapy of Benzyl Isothiocyanate and Cisplatin

As mentioned in Section 2.4, combination therapy of ITCs and cisplatin is expected to have better anti-cancer effects than those of ITCs or cisplatin alone. Indeed, several reports showed that BITC enhances the anti-cancer effects of cisplatin in lung cancer cells (NCI-H596), head and neck squamous cell carcinoma cells (HN12, HN8, and HN30), and leukemia cells (HL-60) [85–87]; there are no similar studies in BC cells, however.

### *2.3. Sulforaphane*

SFN can be found in cruciferous vegetables, such as broccoli, cauliflower, brussel sprouts, cabbage, kale, and kohlrabi [88]. SFN is reported to regulate cancer cell survival via inhibition of cell proliferation and stimulation of apoptosis in a variety of cancers [89,90]. Among the ITC members, SFN has been the most widely investigated regarding its pathological roles and molecular mechanisms both *in vivo* and *in vitro*.

### 2.3.1. In Vitro Studies: Cell Cycle-, Caspase- and Bcl-2-Related Molecules

Regarding the relationships between SFN and cell survival, including cell proliferation, cell cycle, and death, various mechanisms have been suggested. For example, SFN was reported to induce growth arrest and apoptosis in a BC cell line (5637 cells) [91]. Moreover, induction and stimulation of cyclin B1 and Cdk1 were found associated with the anti-proliferative effects of SFN, whereas activation of caspase-3, 8, and 9 and PARP corresponded to its pro-apoptotic effects; these SFN-induced anti-cancer effects are speculated to be regulated via reactive oxygen species (ROS)-dependent mechanisms [91]. Another study showed that SFN treatment suppresses cell viability in a dose-dependent manner and induces apoptosis in T24 human BC cells via regulation of caspase-3, caspase-9, and PARP [92]. Moreover, the SFN-induced apoptosis of BC cells is mediated by dysregulation of mitochondria function, cytochrome *c* release, and Bcl-2-related pathways [92].

Other studies focused on the relationships between cell cycle-related molecules and SFN. For instance, after 10–40 μM SFN treatment for 24 or 48 h, T24 cell viability is significantly suppressed with IC50 values of 26.9 ± 1.12 μM (24 h) or 15.9 ± 0.76 μM (48 h) [93]. Conversely, 20 μM SFN treatment for 24 or 48 h resulted in apoptotic features, such as cell shrinkage, condensed chromatin, and apoptotic bodies, in the same BC cells; increased numbers of apoptotic cells were confirmed by flow cytometry [93]. SFN is also associated with blocking cell cycle progression at G0/G1 phase. In addition to its pro-apoptotic activities, upregulation of the cyclin-dependent kinase inhibitor p27 plays crucial

roles in the 20 μM SFN-induced anti-cancer effects in BC cells, whereas p16 or cyclin D1 expression does not [93]. Thus, regulation of cell cycle-related molecules and mitochondrial function, caspases, and the Bcl-2 protein family represent the molecular mechanisms of SFN-induced anti-proliferative and pro-apoptotic activities in BC cells.

### 2.3.2. In Vitro Studies: Oxidative Stress, Endoplasmic Reticulum Stress, and Growth Factors

As mentioned above, many investigators believe that the anti-cancer effects of SFN in BC are mainly associated with caspase- and mitochondria-related pathways. Nevertheless, there are other cancer-related factors involved. For instance, SFN can inhibit DNA damage induced by chemical carcinogens in BC T24 cells [94]. Moreover, SFN-induced oxidative stress through ROS has been suggested as a key modulator [91,92]. Nuclear factor erythroid 2-related factor-2 (Nrf2) regulation and endoplasmic reticulum (ER) stress are also associated with SFN and carcinogenesis, pathological behavior, and cell survival in UC [28,92]. Notably, these Nrf2 and ER signaling pathways are important factors in the response to oxidative stress and anti-oxidative activities [28,92,95]. A study showed that enhanced insulin-like growth-factor-binding protein-3 (IGFBP-3) and suppressed nuclear factor-kappa B (NF-κB) expression by SFN are associated with the anti-proliferative effect of SFN in the BC cell line BIU87. Interestingly, the authors also found that SFN stimulates apoptosis and cell cycle arrest at the G2/M phase, resulting from IGFBP-3 and NF-κB regulation [96]. As IGFBP-3 and NF-κB are known to possess pro-apoptotic and anti-apoptotic functions, respectively, in various malignancies [97,98], this stimulation of apoptosis by SPN via increased IGFBP-3 and decreased NF-κB levels are in agreement with established findings. Another report on the relationship between SFN-induced anti-cancer effects and growth factors demonstrated that 20 μM SFN leads to a 2.6-, 3.0-, or 3.1-fold increase in the G2/M phase compared with that of controls in three BC cell lines (RT4, J82, and UM-UC-3, respectively) [99]. In addition, SFN induces apoptosis in RT4 and UM-UC-3 cells. Thus, these findings indicate that upregulation of caspase-3/7 and PARP activity and downregulation of survivin, EGFR, and HER2/neu are the underlying molecular mechanisms.

TNF-related apoptosis-inducing ligand (TRAIL) is recognized as an initiator of apoptosis. Its dysregulation has been identified in various malignant cells, including BC [100,101]. As a result, resistance to TRAIL is associated with high malignant potential and worse prognosis for patients with BC [102]. SFN treatment however, has been reported to reverse the pro-apoptotic activity of TRAIL in TRAIL-resistant BC cells [103]; the SFN-induced mechanisms were found associated with apoptosis-related molecules (e.g., caspases, mitochondrial membrane potential, Bid, and death receptor 5) and oxidative stress-related factors (e.g., ROS and Nrf2).

The anti-cancer effects of SFN under hypoxic conditions in BC cell lines have also been reported [88]; in RT112 cells, 20 μM SFN inhibited cancer cell proliferation by 26.1 ± 4.1% and 39.7 ± 5.2% under normoxia and hypoxia, respectively (P < 0.05), with similar results observed for RT4 cells (normoxia, 29.7 ± 4.6%; hypoxia, 48.3 ± 5.2%). Tumor tissues, especially those within the center, are generally under hypoxic conditions due to the oxygen consumption of the tumor to support its growth. Thus, these findings indicate that SFN can suppress cell proliferation under hypoxic conditions in BC with rapid tumor growth compared with that under normoxia and relatively slow growth. Interestingly, the same study also showed that SFN suppresses glycolytic metabolism under hypoxia by decreasing the nuclear translocation of hypoxia-inducible factor-1α, thereby reducing its protein levels [88]. Suppression of glycolytic metabolism in cancer cells is important for inhibiting tumor growth and progression as high glycolytic metabolism leads to increased cancer cell proliferation. Overall, the findings demonstrate that SFN plays several roles in suppressing malignant aggressiveness, such as by decreasing cancer cell proliferation, in BC cells.

2.3.3. In Vitro Studies: Inflammation, Epithelial-to-Mesenchymal Transition, Epigenesis, and Others

In addition to reducing BC cell survival, SFN inhibits malignant aggressiveness by suppressing inflammation, cancer cell invasion, and metastasis. Several studies have shown that SFN downregulates COX-2 expression in BC cells via regulation of p38 mitogen-activated protein kinase (MAPK) and NF-κB [104–106]. Moreover, p38 MARK is positively associated with glutathione transferase and thioredoxin reductase-1—both antioxidant enzymes—following SFN treatment [105]. Furthermore, SFN can inhibit epithelial-to-mesenchymal transition (EMT)—an important mediator of cancer cell invasion and metastasis—via regulation of COX-2/matrix metalloproteinase (MMP)-2, -9/ZEB1, Snail, and miR-200c/ZEB1 in BC cells [106].

SFN was found to inhibit histone status in BC cells, which is associated with reduced levels of histone H1 phosphorylation via modification of histone acetyltransferase and histone deacetylase activity [26]. Changes in histone H1 status were previously reported to be associated with carcinogenesis and prognosis of BC [107]. Based on these findings, SFN is speculated to inhibit carcinogenesis and progression of BC via epigenetic modification [26].

Recently, the physiological and pathological roles of gut microbiota have garnered great interest. Research has shown how they affect systematic metabolism, inflammation, and the immune system, contributing to carcinogenesis, malignant potential, and cancer progression, of which similar findings have been reported in UC [108–110]. Interestingly, SFN was found to normalize gut microbiota dysbiosis by increasing the abundance of *Bacteroides fragilis* and *Clostridium* cluster I in a BBN-induced BC animal model [111], suppressing BBN-induced histological changes, including sub-mucosal capillary growth. While the detailed mechanisms of the anti-carcinogenic function of SFN in this model is not fully clear, normalization of intestinal flora has been shown to repair intestinal barrier dysfunction and injured mucosal epithelium via regulation of tight junction proteins, including ZO-1, claudin-1, occludin, and mucin-2 [111]. Moreover, SFN plays crucial roles in the inflammatory status of this model, as it decreases pro-inflammatory factors such as IL-6 and secretory immunoglobin A, which are increased by carcinogenesis [111]. The authors conclude that these gut microbiota-related beneficial effects of SFN led to its anti-carcinogenic effects in BC via complex mechanisms that involve inflammation and the immune system [111]. A summary of the molecular mechanisms of the anti-cancer effects of SFN is shown in Table 3.


**Table 3.** *In vitro* molecular mechanisms of the anti-cancer effects of sulforaphane.

IGFBP, insulin-like growth-factor-binding protein; NF-κB, nuclear factor-kappa B; HIF, hypoxia-inducible factor; PARP, poly ADP-ribose polymerase; EGFR, epidermal growth factor receptor; HER, human EGFR-related; ROS, reactive oxygen species; ER, endoplasmic reticulum; Nrf2, nuclear factor erythroid 2-related factor; MMP, matrix metalloproteinase; DR5, death receptor 5; EMT, epithelial-to-mesenchymal transition; COX, cyclooxygenase.

#### 2.3.4. In Vivo Studies

*In vivo* studies with the chemical-induced BC animal model have shown that SFN inhibits carcinogenesis, tumor growth, and progression via a complex mechanism that includes prevention of DNA damage [94]. In a murine UM-UC-3 xenograft model, tumor growth rates and tumor weights in the SFN group were found lower than in the control group (not significant and *p* < 0.05, respectively) [51]. Furthermore, this model showed decreases in tumor volumes in SFN-treated mice (12 mg/kg bodyweight for 5 weeks) with an inhibitory rate of 63% via increased caspase-3 and cytochrome *c* expression and decreased survivin expression [113]. In addition to the apoptosis-related pathways, several other molecules have been suggested to be associated with the anti-cancer effects of SFN, based on *in vivo* studies. Thus, further *in vivo* studies are essential for understanding the efficacy and limitations of an SFN-based treatment strategy against BC.

### 2.3.5. Combination Therapy of Sulforaphane and Other Therapeutic Agents

Although several clinical trials on the anti-cancer effects of SFN and broccoli sprout extracts have been performed, the results were unsatisfactory [114,115]; for example, no or minimum effects are detected on serum and tissue biomarkers of patients with prostate and breast cancer. Although a similar clinical trial has not been performed for patients with BC, the clinical effects of SFN monotherapy are also predicted to be unsatisfactory. Therefore, the efficacy of a combination therapy of SFN and other therapeutic agents was investigated in BC cells.

A combination therapy of acetazolamide (AZ; a carbonic anhydrase inhibitor) and SFN showed suppressed proliferative and clonogenic effects and stimulated apoptotic activity via caspase-3 and PARP activation [116]. In addition, the PI3K/Akt signaling pathway was found to play an important role in the anti-cancer effects of this combination therapy. The authors thus conclude that AZ + SFN is a potential therapeutic strategy for BC. Another study examined the effects of two ITCs (AITC + SFN) on the lung cancer cell line A549 [117]; their anti-carcinogenic effects showed higher inhibitory effects on tumor growth and cancer cell migration and greater stimulation of apoptosis compared with that of ATIC or SFN alone. Moreover, oxidative stress, including ROS, is associated with these activities. Although this study was not performed on BC cells, we believe that a combination of different ITCs may be effective for the prevention and treatment of BC. Indeed, in the BC cell line UM-UC-3, pro-apoptotic activity of BITC or PEITC alone is stronger than that of AITC or SFN alone [72]. Moreover, ≥20 μM SFN significantly suppresses cell proliferation in the BC cell line BIU87, whereas 10 μM SFN had no significant effect [96,112]. Therefore, more detailed studies on the combination of various ITC types, dosages, and durations are necessary to identify the most efficacious combination therapy of the ITC members.

Nevertheless, there are potential limitations of SFN-based therapies. Novel immunotherapy strategies, such as immune checkpoint inhibitors, have been recently established as standard therapy for patients with advanced/metastatic BC [118,119]. While we speculate that a combination of immunotherapy and SFN may be useful for the prevention and treatment of BC, a combination of SFN with T cell-mediated cancer immunotherapies is not recommended because SFN can function both as an anti- and pro-carcinogenic factor due to its effects on tumor and immune cells [120].

### *2.4. Phenethyl Isothiocyanate*

Similar to other ITC members, PEITC can suppress carcinogenesis and malignant aggressiveness in various types of malignancies [121]. Suppression of various cancer-promoting characteristics, such as cancer cell proliferation, invasion, and angiogenesis, via regulation of the Bcl-2 protein family, caspases, and matrix metalloproteinases are reported as the potential molecular mechanisms underlying the tumor-suppressive activities of PEITC [121–123].

### 2.4.1. In Vitro Studies

PEITC was shown to have anti-cancer regulatory effects on cancer cell survival and apoptosis in BC cells [124]; PEITC inhibits cell viability in a dose-dependent manner and enhances apoptotic potential, as measured by caspases activities in T24 cells [124]. However, as shown in Table 4, the detailed molecular mechanisms underlying these anti-cancer effects in BC cells are not fully understood. It was reported that PEITC inhibits cell proliferation and stimulates apoptosis in the

human adriamycin (ADM)-resistant bladder carcinoma cell line T24/ADM [123]. Interestingly, this study showed that PEITC increases intracellular drug accumulation potential and DNA topoisomerase II expression, and decreases multidrug resistance-related factors, such as multidrug resistance gene (MDR1), multidrug resistance-associated protein (MRP1), and glutathione S-transferase π [123]. In general, such changes by PEITC lead to increased chemosensitivity. Additionally, the authors clarified the detailed molecular mechanism underlying multidrug resistance reversal potential, which includes downregulation of NF-κB, survivin, Twist, and Akt and upregulation of PTEN and JNK by PEITC. As chemotherapeutic regimens including ADM are the standard therapy for patients with advanced UC [125,126], these findings highlight PEITC as a potential therapeutic agent for BC treatment, especially in patients with drug-resistant BC [123]. While we agree with their conclusion, clinical trials testing this hypothesis have yet to be performed.

**Table 4.** *In vitro* molecular mechanism of the anti-cancer effects of phenethyl isothiocyanate.


Bcl, B-cell lymphoma-2; Bax, Bcl-2-associated X protein; Bak, BCL2-antagonist/killer; NF-κB, nuclear factor-kappa B; JNK, c-Jun N-terminal kinase.

Although BITC has been suggested to suppress cell growth by upregulating miR-99a expression via regulation of the c-Jun/AP-1 pathway in BC [79,80], this pathway plays no significant role in the anti-proliferative effects of PEITC in BC cells [127]. *In vitro* molecular mechanisms of the anti-cancer effects of PEITC are shown in Table 4.

### 2.4.2. In Vivo Studies

PEITC is suggested to play crucial roles in preventing the initiation step of carcinogenesis and inhibiting tumor progression in a variety of malignancies [121]. However, in a chemically BBN-induced BC animal model using male human c-Ha-ras proto-oncogene transgenic rats, microscopic BC is observed in the BBN alone (16 weeks) and BBN (8 weeks) → PEITC (8 weeks) groups; but not in the PEITC (8 weeks) → BBN (8 weeks) group [128]. This finding indicates that PEITC can inhibit the carcinogenic process after initiation. However, a conclusion cannot be drawn due to the limited information on *in vivo* SFN activities in BC.

### 2.4.3. Combination Therapy of Phenethyl Isothiocyanate and Other Therapeutic Agents

We previously introduced a novel treatment strategy that combines AITC and cisplatin for lung cancer cells (Section 2.1.4). Similarly, the efficacy of a combination therapy of PEITC and cisplatin was demonstrated in several studies. For instance, cervical cancer cells (HeLa) treated for 24 h with 5 μM PEITC and 10 μM cisplatin show typical features of apoptosis, such as cell shrinkage, membrane blebbing, and cell detachment, with a 4-fold increase in caspase-3 activity; these significant changes are not observed for either treatment alone [122]. The same study also showed that PEITC increases the pro-apoptotic activity of cisplatin in C33A cervical cancer and MCF-7 breast cancer cells; interestingly, this pro-apoptotic activity is not detected in normal human mammary epithelial MCF-10A cells [122]. Another study on non-small cell lung cancer cells (A549) showed that the percentage cell survival after treatment with AITC (15 μM) or cisplatin (5 μM) alone is 79.2 ± 3.8% and 55.9 ± 3.4%, respectively, whereas cell survival in the combination group is 46.2 ± 2.7% [129]. Notably, when PITC and cisplatin are co-encapsulated in liposomal nanoparticles, A549 cell survival further decreased to 33.3 ± 2.9% [129]. Similar results were obtained in another non-small cell lung cancer cells (H596), where the percentage cell survival after treatment with liposomal-PEITC-cisplatin or free PITC + cisplatin is 55.0 ± 9.5% and 28.6 ± 6.3%, respectively (p < 0.001) [129]. Moreover, the liposomal nanoparticles containing both PEITC and cisplatin have the advantage of increased circulation time in

the bloodstream and accumulation in tumors [130]. Therefore, co-encapsulated PITC and cisplatin in liposomal nanoparticles may be a potential therapeutic strategy for advanced/metastatic UC.

### **3. Carcinogenic Potential of Isothiocyanates**

There is general consensus that all ITC members possess anti-cancer effects in BC cells. However, several studies have also suggested the carcinogenic potential of ITCs in BC. In this section, we will discuss the relationships between BITC, SFN, and PEITC and carcinogenic changes in BC. To our knowledge, AITC has not been shown to promote tumorigenesis and carcinoma in BC; nevertheless, we cannot conclude that AITC has no carcinogenic potential as there is limited information on the biological and pathological effects of AITC in BC.

### *3.1. Carcinogenic Potential of Benzyl Isothiocyanate*

In a two-stage carcinogenesis model, rats treated with BITC and with BBN initiation show neoplastic lesions, including papillary or nodular-hyperplasia (100%), papilloma (38%), and carcinoma (100%); these frequencies are higher than in rats under a basal diet (57%, 5%, and 24%, respectively) [131]. The frequencies of papilloma and carcinoma are also lower than those in rats with initiation + BITC (papilloma = 17% and carcinoma = 0%) and rats without initiation. Therefore, BITC may enhance the carcinogenic process in rats with initiation alone. However, in a BBN-induced BC rat model, oral administration of 10, 100, or 1,000 ppm BBN suppresses carcinogenic pathological changes [83]; moreover, epithelial hyperplasia of the urinary bladder is detected in rats treated with 100 or 1,000 ppm BITC, even without BBN [83]. Furthermore, the same research group showed that these neoplastic changes increase in rats with initiation treatment of 500 ppm BBN and subsequent low dose (25 ppm) BBN exposure, and their frequencies are further increased by additional treatment with 100 and 1000 ppm BITC in a dose-dependent manner [84]. In a similar experiment without initiation treatment, dysplasia, papilloma, and carcinoma were rare, although almost all rats had hyperplasia, except for the control and 100 ppm BITC groups [84]. Thus, BITC may stimulate carcinogenesis in a high-risk population of BBN-induced BC cases [83,84,132].

### *3.2. Carcinogenic Potential of Sulforaphane*

As shown in a previous study, ≥20 μM SFN decreases cell viability and migration of T24 BC cells [112]. However, the study also showed that a low concentration of SFN promotes BC cell proliferation and migration [112], where 1–5 μM SFN or 2.5 and 3.75 SFN increase cell growth to approximately 120–130% and cell migration to 128 and 133% compared with those of control [112]. Thus, a biphasic effect of SFN on cell growth and migration of BC cells was suggested. Mechanistically, activation of autophagy by SFN is speculated to be associated with upregulated cell migration in an *in vivo* study using the autophagy inhibitor 3-methyladenine in T24 cells. The authors also found an enhanced protective effect in conjunction with selenium against free radical-induced cell death. Although this mechanism was confirmed in human hepatocyte cells (HHL-5) and breast cancer cells (MCF-7) rather than in BC cells, the benefits, and risks of SFN have been shown to be dependent on its doses and interactions with the microenvironment, including autophagy and selenium.

### *3.3. Carcinogenic Potential of Phenethyl Isothiocyanate*

In an animal model of dimethylbenzanthracene-induced mammary carcinogenesis, continuous oral administration of 1200 ppm PEITC induces hyperplasia in the urinary bladder [133]. However, another study showed a high frequency of carcinoma (11 of 12 rats; 91.7%) with oral administration of 0.1% PEITC in rats for 48 weeks [134]; the authors thus conclude that PEITC has carcinogenic activities in the rat urinary bladder. By contrast, in a two-stage carcinogenesis model, rats treated with PEITC with BBN initiation exhibit papillary or nodular-hyperplasia (100%), papilloma (24%), and carcinoma (100%). Meanwhile, frequencies of papilloma (17%) and carcinoma (33%) are lower in rats without initiation than in those with initiation [131]. In studies showing the carcinogenic

potential of PEITC according to initiation, specifically in a rat medium-term multi-organ carcinogenesis model, oral treatment with 0.1% PEITC after the initiation period leads to incidences of papillary or nodular-hyperplasia and tumors [135]. However, the authors showed that PEITC provided during the initiation period is not associated with carcinogenic activity [135]. These findings suggest that PEITC may stimulate carcinogenesis of UC during the post-initiation period. Another study, however, demonstrated that PEITC increases the incidences of papillary or nodular hyperplasia, dysplasia, and carcinoma in a dose-dependent manner; thus, > 0.01% PEITC enhances rat urinary bladder carcinogenesis and > 0.05% PEITC has tumorigenic potential. [136]. Collectively, these findings indicate that carcinogenic potential of PEITC administration may be modulated by complex mechanisms that involve timing and dosage.

When 0.1% BITC or 0.1% PEITC is administered in the diet to 6-week-old F344 rats for 1, 2, 3, and 7 days, a significant reduction of urinary pH levels compared to the normal control is detected, starting at day 1 [132]. Similarly, a reduction is detected in the urinary concentration of Na and Cl, whereas K is reduced. The same study also showed that thickness of the urinary bladder urothelium is significantly increased by administration of both BITC and PEITC and that inflammation, vacuolation, erosion, and apoptosis/single cell necrosis occur in the urinary bladder lesion; these morphological changes are not observed in normal control rats [132]. Furthermore, the cell proliferation potential, evaluated by the BrdU labeling index in male rats treated with BITC and female rats with BITC + PEITC, is significantly higher than that of control rats [132]. By contrast, when 0.1% BITC or 0.1% PEITC is administered for 14 days, histopathological simple hyperplasia and papillary/nodular hyperplasia are detected in 100% and 86% and 100% and 60% of the cases, respectively [132]. The authors thus suggest that continuous proliferation of bladder epithelial cells by BITC and PEITC plays important roles in pathological changes, including inflammation and the early stage of carcinogenesis [132]. Taken together, these findings are extremely important for the consideration of ITC treatment strategies, especially BITC and PEITC, for UC. However, we should note the difference in administered levels of ITCs in these studies. Although the mean daily consumption of BITC and PEITC in rats was approximately 80 mg/kg/day [132], these levels do not reflect human physiological conditions, i.e., 30 g of fresh watercress = 7.6 mg of PEITC per person and 0.8 mg from fresh (0.5 mg) and cooked (0.3 mg) Swede-turnips = 0.28 mg/person/day of PEITC [137,138].

### **4. Further Considerations**

As previously mentioned, ITCs exhibit their anti- and pro-carcinogenic activities via complex mechanisms. In this review, we mainly introduced the findings of pre-clinical *in vivo* and *in vitro* studies for easier understanding across the field. However, we would be remiss if we do not mention that other cancer-related factors and signaling molecules affect the biological activities and anti-cancer effects of ITCs in malignancies. These include direct/indirect interactions with Nrf2 and NF-κB, the Nrf2-Kelch-like ECH-associated protein (Keap) 1-antioxidant response element (ARE) signaling pathway, and antioxidant enzymes, such as NAD(P)H quinone reductase (NQO1) and glutathione S transferases (GSTs), through the Nrf2-Keap1-ARE signaling pathway are closely associated with ITC-induced bioactivity [139–141]. We would like to emphasize that further basic research is essential for uncovering the utility and limitations of ITCs in cancer treatment, including BC.

Another important issue to consider is the carcinogenic risk factors of BC, which are affected by a variety of harmful chemical compounds (e.g., cigarette smoke) or physiologically active substances (e.g., sex hormones) [142–144]. With regards to cigarette smoke, cytochrome P450 and phase II detoxification enzymes, such as DOQ1, GSTs, and glucuronosyltransferase inhibit the formation of carcinogenic compounds from tobacco-specific carcinogens, and PEITC modulates such cancer preventive activities [145]. This finding supports the hypothesis that PEITC may suppress the tobacco-related cancer risk in smokers. Indeed, a clinical trial showed that metabolic activation of a tobacco-specific lung carcinogen is significantly suppressed by PEIT treatment [146]. We believe that the cancer risk of BC in smokers may be suppressed by ITCs though similar anti-carcinogenic mechanisms, and thus there is value in performing such clinical trials for BC. Furthermore, the frequency of BC is known to be remarkably higher in men than in women, which is perhaps due to the testosterone-androgen receptor pathways [147]. Interestingly, PEITC was reported to suppress testosterone-induced cancer cell proliferation by downregulating the testosterone-androgen receptor pathway in prostate cancer [148]. Meanwhile, other research has shown that estrogen-mediated pathways are associated with malignant potential and tumor growth of BC [143,144]. In addition, SFN was found to regulate tumor growth of breast cancer cells by modulating estrogen activities [149]. Unfortunately, there is little information on the influence of ITC-mediated sex hormone activity on the malignant potential of BC. Nevertheless, there is a possibility that ITCs affect carcinogenesis and malignant aggressiveness by regulating sex hormones in BC. This highlights the need for designing studies to identify the biological roles of ITCs according to patient background and environment, including occupation, diet, and health habits.

### **5. Conclusions**

In this review, we discussed the anti-cancer effects of ITCs in BC. The research suggests that all ITC members can suppress carcinogenesis, tumor development, and progression *in vivo* and *in vitro*. Furthermore, regulation of cell proliferation, cell cycle, and apoptosis play crucial roles in the ITC-induced anti-cancer effects, and such phenomena are mainly regulated by complex mechanisms involving caspases, Bcl-2 family proteins, and mitochondrial activities. While changes in cancer-related molecules by ITCs may correspond to anti-cancer mechanisms in BC cells, some ITCs may have neoplastic and carcinogenic potential in BC. To clarify this issue, more detailed studies at the molecular level are essential. While there is a possibility that ITC-based treatment strategies can improve prognosis in patients with BC, further clinical trials with well-designed protocols are required to establish the optimal doses and types of ITCs for application in BC treatment [112]. In addition, it would be fruitful to investigate the anti-cancer effects and clinical utility of combination therapies of ITCs and new therapeutic strategies, including immunotherapy and gene therapy, for patients with BC.

**Author Contributions:** Conceptualization, Y.M.; supervision, H.S.; writing—original draft preparation, T.M., Y.M., T.Y., Y.M., A.O., K.M. and K.O. All authors have read and agreed to the published version of the manuscript. The authors declare that the content of this paper has not been published or submitted for publication elsewhere.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Review*
