HMT Inhibits Prostate Cancer Progression by Targeting PAK-1.

Background: Heamatang (HMT) is a classic medicinal formula used in traditional Chinese and Korean medicine; it contains seven distinct components, mainly of herbal origin. HMT is used as an anti-aging remedy, in the treatment of urinary disorders, and to increase energy and vitality. However, the therapeutic applications of this formula have not been evaluated with evidence-based science. Methods: Therefore, we assessed HMT through various in vitro methods, including cell viability assay, uorescence-activated cell sorting assay (FACS), western blotting, migration assay, three-dimensional (3D) cell culture, siRNA-mediated PAK-1 knockdown, and crystal violet assays. Results: HMT decreased PAK-1 expression in PC-3 cells and inhibited cell viability, growth, and motility. The inhibition of cell motility by HMT was correlated with PAK-1-mediated inhibition of Lim domain kinase (LIMK) and colin. HMT induced G1 arrest and apoptosis through the transcriptional regulation of cell cycle regulatory proteins (inhibition of cyclin-dependent kinase (CDK) 4, 6, and cyclin D1) and apoptosis-related proteins (increase in cleaved caspase-3 (c-cas3) and inhibition of poly ADP-ribose polymerase (PARP) and B-cell lymphoma 2 (BCL-2)). Moreover, HMT suppressed P21-activated kinase (PAK-1) expression, leading to the inhibition of AKT activities. Finally, we showed that Decursin was the active ingredient involved in the inhibitory effect of HMT on PAK-1. Conclusion: Our ndings demonstrated that HMT exerts its anticancer inuence through the inhibition of PAK-1 expression. The HMT formula could be applied in various elds, including functional health food and pharmaceutical development. crystal violet staining assay used to determine the anti-proliferative effect of HMT. PC-3 cells were seeded in a plate at a of 1 × 5 cells/ml/well and treated with different concentrations of HMT (0, 50, 100, and 200 μg/ml) for ve days, with daily addition of fresh media and HMT. Then, cells were xed with 2 ml of 1 % glutaraldehyde solution (JUNSEI, Tokyo, Japan) in phosphate-buffered saline (PBS) for 15 min at 37 °C. After washing with PBS, 2 ml of 0.05 % crystal violet (Sigma Aldrich) was added for 30 min to stain cells. Cells were nally washed gently with deionized water. The plates were dried at room temperature overnight. A 70 % ethanol solution was added to each well (2 ml/well) to release the crystal violet, using a rotary shaker for 2 h at room temperature. The O.D. was measured by a microplate reader (Tecan, Sunrise, Switzerland) at 570 nm, with a reference lter at 405 nm. expression levels and inducing the expression of cleaved Caspase-3. Interestingly, HMT treatment and PAK-1 knockdown by siRNA showed a similar phenotype in PC-3 cells. Moreover, we identied that the inhibition of PAK-1 by HMT treatment is dependent on Decursin. Our ndings demonstrated that HMT inhibits cell motility and growth and induces apoptosis by inhibiting the PAK-1/AKT and the PAK-1/LIMK1/2/Colin signaling pathway. Collectively, these data suggest that HMT could be used as a starting point for the development of functional health foods and medicine.

PAK-1 is a major downstream effector of the Rho-family GTPase Cdc42 and Rac1 and is involved in the regulation of cell morphology and motility [5].
Cancer cell migration is a fundamental process in solid tumors formation and is required for metastasis formation [6]. The formation of cancer metastasis, which includes cancer cell migration and invasion, involves changes in cytoskeletal signaling pathways, increased motility, and enhanced cell survival. Thus, the actin cytoskeleton is an important factor in tumor cell migration and invasion [7]. It has been reported that PAKs regulate the actin cytoskeleton during cell motility and invasion [8]. The phosphorylation of both LIMK and co lin is greatly enhanced in the presence of active PAK [9][10][11][12]. Furthermore, the expression and activity of LIMK are higher in invasive breast and prostate cancer cell lines than in less invasive lines [13]. In addition, PAK-1 plays an essential role in cell growth, adhesion, migration, and survival in colorectal cancer [14] and lung adenocarcinoma [15], through the activation of AKT, ERK, and β-catenin.
The use of Complementary and alternative medicine (CAM) treatment methods, including traditional Korean medicine, has shown a steady rise among cancer patients. Many patients seek CAM therapeutic options to mitigate the side effects of chemotherapy and radiation therapy [16,17]. However, further studies are required to examine the e cacy and safety of CAM in anticancer therapies. Heamatang (HMT) is a traditional medicine that was described in the Compendium of Materia Medica (Bancao Gamgmu), a Chinese herbalogical and pharmaceutical textbook written by Li Shizhen during the Ming dynasty reign [18]. According to Bancao Gamgmu, HMT disperses hard masses caused by accumulation and assemblage of blood clot or waste discharge under de cient and excessive conditions over a long time. Thus, HMT could be applied to cancer treatment. However, there is a lack of scienti c evidence regarding its clinical e cacy, as the therapeutic effects of this classic prescription have not been evaluated using conventional scienti c methods. HMT is constituted by combining seahorse, and six herbal components, including Rheum palmatum, Pharbitidis Nill Choisy, Citri unshius Markovich, Inula helnium, Croton tiglium Linne, and Angelica gigas Nakai. HMT has not been employed in modern medicine because the use of Hippocampus species was prohibited on account of the originating species being endangered. However, we were able to produce HMT in-house thanks to the farming of Hippocampus abdominalis together with its approval for the use as food material in Korea. Therefore, in this study we evaluated the impact of HMT in prostate cancer therapy, using conventional scienti c methods.

Preparation of HMT
The seven ingredients and their proportion (w/w) were as follows: Hippocampus abdominalis (Seahorse Australia, Beauty Point, TAS, Australia), 16 (http://mpns.science.kew.org). The preparation method is the following: dried and pulverized medicinal herbs were mixed and soaked in 50 % ethanol (1L x 3 changes) at room temperature for three days. The extract was ltered using Whatman lter paper, evaporated (rotary evaporator, model NE-1, Japan) and lyophilized (freeze dryer, Lioalfa-6, Telstar, Terrassa, Spain) to produce 15 g of powder.

Crystal Violet Staining Assay, Cell Growth Assay
The crystal violet staining assay was used to determine the anti-proliferative effect of HMT. PC-3 cells were seeded in a 6 well plate at a concentration of 1 × 10 5 cells/ml/well and treated with different concentrations of HMT (0, 50, 100, and 200 μg/ml) for ve days, with daily addition of fresh media and HMT. Then, cells were xed with 2 ml of 1 % glutaraldehyde solution (JUNSEI, Tokyo, Japan) in phosphate-buffered saline (PBS) for 15 min at 37 °C. After washing with PBS, 2 ml of 0.05 % crystal violet (Sigma Aldrich) was added for 30 min to stain cells. Cells were nally washed gently with deionized water. The plates were dried at room temperature overnight. A 70 % ethanol solution was added to each well (2 ml/well) to release the crystal violet, using a rotary shaker for 2 h at room temperature. The O.D.
was measured by a microplate reader (Tecan, Sunrise, Switzerland) at 570 nm, with a reference lter at 405 nm.

siRNA Transfection
The PAK-1 siRNA and control siRNA were purchased from Bioneer. To transfect the siRNAs, PC-3 cells were seeded at a density of 5 × 10 4 cells per well in a 6-well plate. Cells were transfected using 100 nM of either the PAK-1 siRNA or the control siRNA with a siRNA transfection reagent (Polyplus) for 48 h. After treatment, cells were stimulated with HMT and then analyzed with western blotting assay and stained with crystal violet. The O.D. was measured by a microplate reader (Tecan, Sunrise, Switzerland) at 570 nm, with a reference lter at 405 nm.

3D Culture Tumor Organoids
For the generation of the PC-3 tumor organoids, cells were seeded into 96-well round-bottom ultra-low attachment plates (Corning) at 2000 viable cells per well. The PC-3 spheroids were grown in RPMI medium with 10 % FBS. The plates were incubated for 5 days at 37 °C, during which they formed spheroids. Then, the spheroids were treated with 200 μg/ml HMT for 48 h. For the apoptosis analysis, 2 μM CELL Event (Invitrogen, USA) was added to each well for 1 h. Pictures were obtained using a uorescence microscope (Nikon, Tokyo, Japan). Fluorescence intensity of caspase-3/7 was evaluated using the ImageJ software. For the calculation of uorescence intensity with ImageJ, two images of each well were analyzed and their average intensity was used for the statistical analysis.

TLC and HPLC Analysis
Decursin standard and HMT were spotted on Silica gel 60 thin-layer plates (Merck, Darmstadt, Germany).
After development in isopropyl alcohol: ethyl: acetate: water (3/1/1, v/v/v), the TLC plate was dried and visualized by dipping in a solution containing 0.3 % (w/v) N-(1-naphthyl)-ethylenediamine and 5 % (v/v) H 2 SO 4 in methanol and was heated at 110 °C for 10 min. We then performed a high-performance liquid chromatography (HPLC) on a Luna C18(2) 100 Å reverse phase column (250 mm×4.6 mm, 5 μm, phenomenex, Inc., Korea) connected to an Agilent Technologies 1100 series system, with a detector at 329 nm at 30 °C and using a mobile phase of 100 % ethanol: 0.1 % formic acid in water (7/3, v/v) under isocratic mode with a ow rate of 1 ml/min. The chromatography peak was identi ed by comparing the retention time of the sample with the reference standard.

HMT Suppresses PAK-1
The PC-3 line has a high metastatic potential among prostate cancer cell lines. We thus tested whether HMT treatment could reduce PAK-1 expression in PC-3 cells by western blotting assay. Indeed, HMT inhibited the PAK-1 protein level in PC-3 cells (Fig. 1A).

HMT Inhibits Cell Growth
We then examined the cytotoxic effect of HMT in several prostate cancer cell lines. We performed a cell viability assay 24 h after HMT treatment in the PC-3, DU-145, and LNCaP lines. As shown in Fig. 1B, HMT treatment reduced cell viability in PC-3, DU-145, and LNCaP cells in a dose-dependent manner. The treatment with 31.3 μg/ml HMT reduced prostate cancer viability by 12.2 % (PC-3), 13.5 % (DU-145), and 53.9 % (LNCaP). LNCaP cells were the most sensitive to HMT (Fig. 1B).
We then performed a cell growth assay to assess the impact of HMT treatment in the long-term (4d) growth of prostate cancer cells. As shown in Fig. 1C, HMT signi cantly suppressed cell growth in a dosedependent manner. HMT suppressed cell growth, decreasing cell growth by 28 %, 46 %, and79 % at 50, 100, and 200 μg/ml of HMT, respectively (Fig. 1C).

HMT Inhibits Cell Migration and Invasion by Inhibiting PAK-1 Pathway
To investigate whether HMT could suppress prostate cancer cell motility without exhibiting cell toxicity, we used a wound-healing and Matrigel-coated membrane invasion assay. As shown in Fig. 2A and 2B, in the Matrigel-coated invasion assay, HMT inhibited cell invasion by 50 % compared to the untreated control. Moreover, in the wound healing assay, HMT treatment decreased serum-induced cell migration by 57 % compared to the untreated control. As shown in Fig. 1B, the inhibitory effect of cell motility by HMT was not linked to cell toxicity. Notably, 62.5 μg/ml HMT suppressed cell viability by 15 % in the PC-3 cell lines compared to the control (Fig. 1B). To con rm that the inhibitory effect of HMT in cancer cell migration and invasion were correlated to PAK-1 regulation, we performed western blotting to measure the protein levels changes of PAK-1 and PAK-1-regulated cell motility proteins following HMT treatment in PC-3 cells. A concentration of 50 μg/ml HMT attenuated PAK-1, AKT, LIMK, and co lin protein levels (Fig.  2C).

HMT Induces G1 and SubG1 Arrest
To evaluate whether HMT could modulate cell cycle progression, cell cycle analysis was performed by treating PC-3 cells with high concentrations of HMT (10,200, and 400 μg/ml) for 24 h. As shown in Fig.  3A, HMT gradually increased G1 and sub-G1 duration in a dose-dependent manner compared to those in the non-treated cells. Particularly, the increase in sub-G1 arrested cells treated with 400 μg/ml HMT was 76 % (Fig. 3A).
To con rm the presence of HMT-mediated apoptotic cells, we stained PC-3 cells treated with 200 μg/ml HMT for 24 h with the Cell Event Caspase-3/7 detection dye. As shown in Fig. 3C, we detected no cas3/7 uorescent cells in the control group. On the other hand, HMT-treated cells showed green uorescence in almost all cells.
We next assessed whether HMT-mediated G1 and sub-G1 arrest (apoptosis) were dependent on changes in G1-regulating and apoptotic-related proteins. HMT treatment decreased the levels of G1-regulated proteins (Cyclin D1 and CDK4 and 6) and increased apoptosis through the inhibition of BCL-2 and PARP and the increase of cleaved Caspase-3 (Fig. 3D).

PAK-1 Mediates HMT-Induced G1 Arrest and Apoptosis, and Suppression of Cell Proliferation and Cell Motility
Next, we assessed whether the anticancer effect of HMT is dependent on PAK-1. Upon knockdown of PAK-1 by siRNA, we observed a reduction of the expression of the G1 regulatory protein cyclin D1, and the induction of the apoptotic protein cleaved Caspase-3 (Fig. 4A). Moreover, the -1 knockdown attenuated the expression of LIMK1/2 and co lin. The expression pro le of PAK-1 knocked down cells was similar to that observed in HMT-treated cells (Fig. 4A). Moreover, PAK-1 knockdown enhanced HMT-mediated inhibition of cell growth (Fig. 4B). In addition, HMT treatment signi cantly suppressed PAK-1 expression in two prostate cancer cell lines, LNCaP and DU145 (Fig. 4C).

HMT Reduces Tumor Spheroid Viability
We used the PC-3 tumor spheroid model further to assess the effect of HMT treatment on tumor growth. This three-dimensional culture model mimics some aspects of the in vivo tumor organization and is better suited to study the response of cancer cells to a drug. As shown in Fig. 5A, the cell viability of the PC-3 spheroids, as measured by WST-8, was reduced by 41 % following the addition of 200 μg/ml of HMT, similarly to what was observed in 2D cultures (Fig. 4A). To measure the effect of HMT on apoptosis in the 3D PC-3 cells model, we used Cell Event, a uorogenic Caspase-3/7 substrate: following HMT treatment, we readily detected apoptosis also in this model (Fig. 5B).

HMT Contains Decursin that Inhibits PAK-1
To identify the active compounds in HMT correlated to the anticancer activity, we determined the presence in HMT of the main herbal compounds (Naringin, Decursin, Catechin, Gallic acid, and Epicatechin) using TLC (data not shown). This preliminary TLC analysis (254 nm) revealed the presence of Decursin, as shown in Fig. 6A.
We then performed HPLC analysis to calculate the retention time of Decursin (5.014 min) and estimated its abundance in the HMT preparation (23.05 mg/g) (Fig. 6B). We then evaluated the effect of pure Decursin on -1 expression. At a concentration of 10 μM, Decursin signi cantly suppressed PAK-1 expression (Fig. 6C).

Discussion
Since PAKs were discovered in 1994, several studies demonstrated their crucial roles in numerous cellular processes, such as cell cycle progression and cell survival [19]. PAK-1 belongs to the Group I of the PAK family and was identi ed for the rst time in a screening for proteins that interacted with GTP-bound Rac [20]. PAK-1 is overexpressed in a variety of cancers, including prostate cancer [21]. PAK-1 was recently shown to be a valid therapeutic target for cancer treatment [22]. Since then, several PAK-1 inhibitors have been developed for use as biological markers and therapeutic agents [23].
Recent studies investigating the antitumor effect of natural products suggested that they may lead to promising alternative therapy for the treatment of cancer. Traditional Chinese or Korean medicine is widely accepted as an alternative treatment for cancer [24][25][26]. Furthermore, research on traditional Chinese and Korean medicinal and herbal formulations has been recognized internationally by medical researchers [27], and several studies elucidated the molecular and cellular mechanisms of herbal medicine-derived phytochemicals in the context of cancer research [28][29][30]. In this study, we investigated the effect of oriental medicinal herbs on PAK-1 signaling. Previously, we have reported the effect of PAK-1 inhibitor EOPK (Essential oil of Pinus koraiensis) in HCT 116 cells [31]. Now, we identi ed a second PAK-1 inhibitor isolated from HMT.
Recent studies have shown that there are several possible links between PAK and PI3K and Raf-MAPK pathways [4,[31][32][33]. To con rm the involvement of PAK-1 in the HMT-mediated anticancer effect, we evaluated expression changes of proteins involved in the PAK-1 pathway by knocking down PAK-1 with siRNA. This knockdown enhanced the HMT-mediated anticancer effects by inhibiting the PAK-1/AKT pathway. Consistently with our data, PAK-1 inhibition using shRNA or siRNA strategies induced apoptosis and CDK4/6 inhibition [34][35][36].
HMT is mainly prepared using extracts from Hippocampus abdominalis, Angelica gigas Nakai, Rhem palmatum, and Critri unshius Markovich. The Hippocampus extract contains steroids and fatty acids and possesses various pharmacological properties including anti-tumor effects [37,38] and anti-prostatic hyperplasia activity [39,40]. Rhem palmatum contains emodin, that has been reported to target the AR directly and induce apoptosis [41,42]. Moreover, emodin has been reported that inhibits tumor cell migration through suppression of PI3-K-cdc42/Rac1 pathway in MDA-MB-231 cells [43]. Rhem palmatum contains catechin and epicatechin. Green tea catechins have been reported to possess anticancer effects by many researchers [44][45][46][47]. The therapeutic mechanisms of catechins, especially in prostate cancer, include direct action on cancer cells and indirect action on tumor-associated in ammation [48]. Critri unshius Markovich contains hesperidin and narinagin. According to a recent research, naringin inhibits cell viability and induces apoptosis in PC-3, LNCaP, and DU145 cells. Furthermore, naringin synergistically increases the effect of paclitaxel in the treatment of prostate cancer cells [49]. Korean Angelica gigas Nakai is a major medicinal herb. Traditionally, its dried root has been used to treat anemia, pain, infection, and articular rheumatism in Korea, most often by boiling the roots in water. TLC and HPLC data show that HMT contains Decursin (Fig. 6A and B). Decursin is a major chemical component of Angelica gigas extract and has been previously been associated with antitumor effects in prostate cancer [50][51][52][53]. Moreover, Decursin decreases cell proliferation and angiogenesis and increases apoptosis in vitro and in PC-3 and DU-145 xenograft models [52,54]. By using TLC, we did not nd the presence of other compounds (Naringin, Emodin, Catechin, Gallic acid, and Epi-catechin). However, further analyses are necessary to identify the presence of other compounds, since the anticancer activity of these components may contribute to the total HMT anti-tumor activity.
Three-dimensional (3D) growth of immortalized cell lines or primary cells is regarded as a more physiological model to perform in vitro screening, as 3D cell cultures possess several in vivo features of tumor organization [55,56]. We evaluated the anticancer-effect of HMT in a 3D culture growth model. As shown in Fig. 5B, the PC-3 tumor organoids were bumpy, not showing the typically spheroid and uniform morphology. Thus, the PC-3 line has high metastatic potential among prostate cancer cell lines. PC-3 cells lack cell-cell adhesion molecules that are related to cell motility, migration, and invasion, making it di cult to form spheroid tumor formation in the absence of a synthetic matrix, such as ECM (extracellular matrix) components. Also, previous studies reported that PC-3cells have no or low levels of E-cadherin and alpha-catenin [57]. As shown in Fig. 5A, the organoids of the control group were brighter than those of the HMT-treated group in bright eld pictures. On the other hand, the spheroids of the HMTtreated group were darker than those of the control group in the whole area. Consistently, the Cell Eventpositive spheroids showed a dark section, consistent with apoptosis induction [58][59][60].

Conclusions
In summary, we demonstrated that, in PC-3 cells, a non-toxic dose of HMT decreased cell migration, cell invasion, and the expression of PAK-1, AKT, LIMK1/2, and Co lin. HMT reduces cell proliferation by arresting cells in the G1 phase, through the reduction of CDK4/6 and cyclin D1 expression levels. HMT also induces apoptosis through sub-G1 arrest by decreasing PARP and BCL-2 expression levels and inducing the expression of cleaved Caspase-3. Interestingly, HMT treatment and PAK-1 knockdown by siRNA showed a similar phenotype in PC-3 cells. Moreover, we identi ed that the inhibition of PAK-1 by HMT treatment is dependent on Decursin. Our ndings demonstrated that HMT inhibits cell motility and growth and induces apoptosis by inhibiting the PAK-1/AKT and the PAK-1/LIMK1/2/Co lin signaling pathway. Collectively, these data suggest that HMT could be used as a starting point for the development of functional health foods and medicine. growth, for long term treatment of HMT was carried out by cell growth-crystal violet assay. HMT (50, 100, and 200 μg/ml) treated to PC-3cells for 4 days. The cells were stained with crystal violet and were photographed and resolved in 70% EtOH, and the absorbance was measured using a microplate reader.

Figure 4
Effect of PAK-1 siRNA on cell migration, proliferation, and apoptosis-related markers in HMT treated PC-3 cells. PAK-1 siRNA and HMT treated PC-3 cell lysates were prepared and subjected to western blotting (A) with antibodies against PAK-1, AKT, p-AKT, LIMK1/2, Co lin, CyclinD1, C-Cas3 and β-actin, and (B) cell growth was assayed by crystal violet staining. The stained cells were photographed and resolved in 70% EtOH, and the absorbance was measured using a microplate reader. Data represent mean ± SD. *p<0.05, **p<0.01 and ***p<0.001 compared with control. (C) HMT-treated PAK-1 expression in LNCaP and DU145 cells by Western Blotting.