Acetylshikonin suppressed growth of colorectal tumour tissue and cells by inhibiting the intracellular kinase, T‐lymphokine‐activated killer cell‐originated protein kinase.

Background and Purpose Overexpression or aberrant activation of the T‐lymphokine‐activated killer cell‐originated protein kinase (TOPK) promotes gene expression and growth of solid tumours, implying that TOPK would be a rational target in developing novel anticancer drugs. Acetylshikonin, a diterpenoid compound isolated from Lithospermum erythrorhizon root, exerts a range of biological activities. Here we have investigated whether acetylshikonin, by acting as an inhibitor of TOPK, can attenuate the proliferation of colorectal cancer cells and the growth of patient‐derived tumours, in vitro and in vivo. Experimental Approach Targets of acetylshikonin, were identified using kinase profiling analysis, kinetic/binding assay, and computational docking analysis and knock‐down techniques. Effects of acetylshikonin on colorectal cancer growth and the underlying mechanisms were evaluated in cell proliferation assays, propidium iodide and annexin‐V staining analyses and western blots. Patient‐derived tumour xenografts in mice (PDX) and immunohistochemistry were used to assess anti‐tumour effects of acetylshikonin. Key Results Acetylshikonin directly inhibited TOPK activity, interacting with the ATP‐binding pocket of TOPK. Acetylshikonin suppressed cell proliferation by inducing cell cycle arrest at the G1 phase, stimulated apoptosis, and increased the expression of apoptotic biomarkers in colorectal cancer cell lines. Mechanistically, acetylshikonin diminished the phosphorylation and activation of TOPK signalling. Furthermore, acetylshikonin decreased the volume of PDX tumours and reduced the expression of TOPK signalling pathway in xenograft tumours. Conclusion and Implications Acetylshikonin suppressed growth of colorectal cancer cells by attenuating TOPK signalling. Targeted inhibition of TOPK by acetylshikonin might be a promising new approach to the treatment of colorectal cancer.

acetylshikonin diminished the phosphorylation and activation of TOPK signalling. Furthermore, acetylshikonin decreased the volume of PDX tumours and reduced the expression of TOPK signalling pathway in xenograft tumours.
Conclusion and Implications: Acetylshikonin suppressed growth of colorectal cancer cells by attenuating TOPK signalling. Targeted inhibition of TOPK by acetylshikonin might be a promising new approach to the treatment of colorectal cancer.

| INTRODUCTION
Colorectal cancer is one of the most common causes of cancer-related death. It comprises a substantial proportion of the global burden of cancer morbidity and mortality, being the fourth most common cause of cancer mortality, accounting for 600,000 deaths annually (Rossi, Anwar, Usman, Keshavarzian, & Bishehsari, 2018). Colorectal cancer is the third most commonly diagnosed cancer among men and the second most diagnosed among women worldwide (Kolligs, 2016;Roswall & Weiderpass, 2015). New cases diagnosed each year in the United States are estimated to be around 95,520 which results in about 50,260 patients deaths (Siegel et al., 2017). Although there have been remarkable progress in developing anticancer therapies including recently approved immune check point inhibitors (Davide et al., 2019), the incidence of and mortality from colorectal cancer is still alarming. Because of the heterogenous nature of colorectal cancer, interventions in a range of oncogenic signalling pathways would be a rational approach for developing new therapies. One of the emerging oncogenic signalling molecules is an intracellular protein kinase, the T-lymphokine-activated killer cell-originated protein kinase (TOPK; also known as PDZ binding kinase, PBK), which has been implicated in the development and progression of gastric (Ohashi et al., 2017) and ovarian cancers (Ikeda et al., 2016), oesophageal squamous cell carcinomas (Ohashi et al., 2016), and colorectal cancer (Zlobec et al., 2010).
There have been few reports of synthesis and functional studies of TOPK inhibitors, such as HI-TOPK-032  and OTS964 (Sugimori et al., 2017). Whereas HI-TOPK-032 reduced the growth of colon cancer xenograft tumours in mice , OTS964 inhibited the size of glioma stem cells tumour spheres in vitro (Sugimori et al., 2017). However, the latter study also demonstrated that the surviving glioma stem cells start to regrow as tumour spheres, thus limiting the therapeutic value of OTS964. Moreover, OTS964 showed substantial adverse haematological reactions in an in vivo xenograft study (Matsuo et al., 2014). Hu et al. (2019) recently reported the synthesis of a series of 1-phenyl phenanthridin-6(5H)one compounds as TOPK inhibitors, which reduced the growth of colorectal tumour growth in a xenograft mouse model. These studies What is already known • The intracellular protein kinase, TOPK/PBK, is highly expressed in colorectal cancer cells.

What this study adds
• Acetylshikonin inhibited colorectal cancer growth dependent on the expression of TOPK/PBK.

What is the clinical significance
• Acetylshikonin could provide a new approach to treatments for colorectal cancer.
suggest that TOPK is a valid drug target for developing anticancer therapies.
In the present study, we have attempted to evaluate the potential of developing acetylshikonin, a major biologically active compound present in Lithospermum erythrorhizon root (Cho, Paik, & Hahn, 1999;Rajasekar et al., 2012), as a TOPK inhibitor and assess its anti-cancer effects in cultureds of colorectal cancer cells and in patient-derived xenograft (PDX) tumour models in mice. Acetylshikonin reportedly inhibits human pancreatic cancer cell proliferation through inhibition of NF-κB activity, attenuates HepG2 hepatoma cell growth by suppressing CYP2J2, and reduces obesity and hepatic steatosis in db/ db mice (Cho & Choi, 2015;Gwon, Ahn, Chung, Moon, & Ha, 2012;Park et al., 2017). These findings led us to examine whether acetylshikionin could inhibit colorectal tumour growth and to elucidate its underlying mechanisms. Here, we report that acetylshikonin directly interacts with TOPK and inhibits TOPK kinase activity, resulting in reduced proliferation of colon cancer cells and diminished growth of PDX tumours in mice. Our study suggests that acetylshikonin, as an inhibitor of TOPK, may be a potential candidate for clinical development as an anticancer therapy for colorectal cancer. McCoy's 5A , or L-15 (SW 620, SW 480) containing penicillin (100 unitsÁml −1 ), streptomycin (100 μgÁml −1 ), and 10% FBS (Biological Industries, Kibbutz Beit-Haemek, Israel). The cells were maintained at 5% CO 2 , 37 C in a humidified atmosphere. All cells were cytogenetically tested and authenticated before being frozen.
Each vial of frozen cells was thawed and maintained in culture for a maximum of 8 weeks.

| Computational modelling
To further confirm that acetylshikonin can bind with TOPK, we performed in silico docking using the Schrödinger Suite 2017 software programs (Schrödinger, 2017). The sequence of TOPK was downloaded from the National Center for Biotechnology Information (GI: 83305809). The TOPK crystal structure was built with prime followed by refining and minimizing loops in the binding site, and then it was prepared under the standard procedures of the Protein Preparation Wizard. Hydrogen atoms were added consistent with a pH of 7, and all water molecules were removed. The TOPK ATPbinding site-based receptor grid was generated for docking.
Acetylshikonin was prepared for docking by default parameters using the LigPrep program. Then, the docking of acetylshikonin with TOPK was accomplished with default parameters under the extra precision (XP) mode using the program Glide. Then, TOPK modelling structure was refining and minimizing loops in the binding site. When the docking was performed, usually several docking models were generated. Herein, we based on the docking score and ligand interaction diagram to choosing best docked representative structure for our final docking model.

| Cell proliferation assay
The cells were seeded (2 × 10 3 cells per well for HCT-116 and HCT-15, 4 × 10 3 cells per well for SW 620, and 1 × 10 3 cells per well for DLD-1) in 96-well plates and incubated for 36 hr and then treated with different doses of acetylshikonin or vehicle. After incubation for 24, 48, or 72 hr, cell proliferation was measured by MTT assay. For anchorage-independent cell growth assessment, the cells (8 × 10 3 per well) suspended in 10% maintenance media were added to 0.3% agar with vehicle, 0.3-, 0.6-, or 1.25-μM acetylshikonin, in a top layer over a base layer of 0.5% agar with vehicle, 0.3-, 0.6-, or 1.25-μM acetylshikonin. The cultures were maintained at 37 C in a 5% CO 2 incubator for 3 weeks, and then colonies were visualized under a microscope and counted using the Image-Pro Plus software program (Media Cybernetics, Rockville, MD, USA, Version 6.0, RRID: SCR_007369).

| Analysis of cell cycle and apoptosis
The cells (2 × 10 5 cells) were seeded in 60-mm dishes and treated with 0-, 5-, or 10-μM acetylshikonin for 24 hr. For cell cycle analysis, the cells were fixed in 70% ethanol and stored at −20 C for 24 hr.
After staining with annexin-V for apoptosis or propidium iodide for cell cycle assessment, the cells were analysed using a BD FACSCalibur Flow Cytometer (BD Biosciences, San Jose, CA, USA).
After centrifugation at 18,000 x g for 20 min, the supernatant fractions were harvested as total cellular protein extracts. Determination

| Reporter gene activity assay
Transient transfection was conducted using the Simple-Fect Transfection Reagent, and the luciferase reporter gene activity assays were performed according to the instructions of the manufacturer (Promega, Madison, WI, USA). Briefly, the cells (2 × 10 4 ) were seeded the day before transfection into 24-well culture plates. The cells were co-transfected with the NF-κΒ-luc or pGL-3-luc firefly reporter plasmid and the pRL-SV control Renilla reporter plasmid. Following incubation for 24 hr, the cells were treated with vehicle or acetylshikonin for an additional 24 hr and then harvested in a lysis buffer. Firefly and Renilla luciferase activities were assessed using the substrates provided in the reporter assay system (Promega, Cat#E2810). The Renilla luciferase activity was used for normalization of transfection efficiency.

| PDX mouse model
All animal care and experimental procedures complied with, and were approved by, the Ethics Committee of Zhengzhou University (Zhengzhou, Henan, China). Six-to eight-week-old female NOD.CB17-Prkdcscid/NcrCrl mice (Vital River Labs, Beijing, China) with body weight 18-20 g were used for these experiments. Animals were housed in a specific pathogen-free facility using IVC level cages, bedding material made of corncob granules, housing five mice per cage.
The mice were maintained on a 12-hr light/dark cycle with food and water available ad libitum, at a temperature of 22 ± 3 C and 40-70% relative humidity. Animal studies are reported in compliance with the ARRIVE guidelines (Kilkenny et al., 2010) and with the recommendations made by the British Journal of Pharmacology.
The PDX mouse model has been widely used in preclinical studies to identify therapeutic targets, including specific molecules and molecular interactions, and to serve as a guide for clinical treatment of cancer (Lai et al., 2017). The PDX tumour samples were obtained from three different patients (HJG41, HJG175, and HJG152; see Figure S5B for details) with permission from the Ethical Committee of China-US (Henan) Hormel Cancer Institute and with full informed consent of the patients. The PDX tumour mass (100-130 mg per mouse) was subcutaneously implanted into the back of SCID mice. When tumours reached an average volume of~100 mm 3 , mice were divided into three treatment groups by randomization and blinding methods for further experimentation as follows: for HJG41 and HJG175 cases, vehicle group (n = 5 for HJG41, n = 8 for HJG152); (2) 60 mgÁkg −1 of acetylshikonin (n = 5 for HJG41, n = 8 for HJG152); and (3) 120 mgÁkg −1 of acetylshikonin (n = 5 for HJG41, n = 8 for HJG152). For HJG 175 cases, vehicle group (n = 10); (2) 80 mgÁkg −1 of acetylshikonin (n = 10); and (3) 160 mgÁkg −1 of acetylshikonin (n = 10). Tumour volume was calculated from measurements of three diameters of the individual tumour base using the following formula: Acetylshikonin was administered by gavage once a day until tumours reached~1.0 cm 3 total volume; at which time mice were killed by an overdose of pentobarbital (4%, i.p.), before tumour collection.

| Immunohistochemical (IHC) analysis
Subcutaneous tumours were collected from the mice and fixed, and paraffin-embedded sections (5 μm) were prepared for IHC analysis.
After antigen unmasking, the sections were blocked with 5% goat serum and incubated at 4 C overnight with antibodies rabbit anti- (1:100), and rabbit anti-phosphorylated c-Jun (pc-Jun) (1:100). After incubation with a rabbit secondary antibody, DAB (3,3 0 -diaminobenzidine) staining was used following the manufacturer's instructions to visualize the protein targets. Sectioned tissues were counterstained with haematoxylin, dehydrated through a graded series of alcohol into xylene, and mounted under glass coverslips.
Images were obtained using OLYMPUS IMAGING BX43. The fluorescence intensity was quantified using Image-Pro Plus software (Version 6.0, RRID:SCR_007369).

| Data and statistical analysis
The design and analysis of this study complies with the recommendations of the British Journal of Pharmacology on experimental design (Curtis et al., 2018). Animals were randomly allocated to the experimental groups. Data collection and evaluation of all experiments were performed blindly of the group identity. Statistical analysis was carried out using the software GraphPad Prism (v7, GraphPad Software, USA, RRID:SCR_002798). Differences among multiple groups were tested using one-way ANOVA followed by Dunnett's post hoc comparisons.
Post hoc tests were conducted only if F was significant and there was no variance inhomogeneity. A value of P < .05 was used as the criterion for statistical significance, and the data are shown as mean values ± SD.

| RESULTS
3.1 | Acetylshikonin is a novel inhibitor of TOPK but has no effect on MEK1 To examine the effects of acetylshikonin on TOPK activity, we conducted a TOPK kinase assay. Acetylshikonin inhibited the kinase activity of TOPK but had no effect on the kinase activity of MEK1, which is another subfamily of the MAPKKs (Figure 1a,b). The direct binding between acetylshikonin and TOPK was revealed by ex vivo and in vitro pull-down assays (Figure 1c,d). Acetylshikonin was incubated with SW 620 cell lysate ex vivo or with a human recombinant active TOPK protein in vitro. To confirm whether TOPK is a major target for acetylshikonin, we performed a kinase profiling assay.
Kinase profiling assay results showed that acetylshikonin (20 μM) inhibited the kinase activity of Aurora A, Aurora B, and c-Src but had no effect on the kinase activity of PDK1, Akt1, JNK1, or ERK1 ( Figure S1A). We then conducted kinase assays to further confirm the inhibitory effects of acetylshikonin against Aurora A, Aurora B, or c-Src activity ( Figure S1B-D). The kinase assay results showed F I G U R E 1 TOPK is a potential target of acetylshikonin.  Figure S4).

| Acetylshikonin inhibits the growth of TOPKpositive PDX tumours in SCID mice
To examine the anti-tumour activity of acetylshikonin in vivo, we used three PDX models, HJG41, HJG175, and HJG152, which exhibited a range of levels of TOPK (Figures 7, S5, and S6). The PDX tumour mass was implanted into SCID mice, and then vehicle or acetylshikonin (60 or 120 mgÁkg −1 body weight for HJG41 and HJG152, and 80 or 160 mgÁkg −1 body weight for HJG175) was administered by oral gavage once a day for 50, 88, or 46 days for HJG41, HJG152, and HJG175, respectively. The results showed that treatment of HJG41 or HJG175 PDX tumour-bearing mice with acetylshikonin at a dose of 120 or 160 mgÁkg −1 , respectively, significantly reduced tumour volume and weight compared to the vehicle-treated group (Figures 7a,b and S6A, B), without changing body weight (Figures 7c and S6C).
However, tumour growth of HJG152, which had low expression of TOPK, was not affected by treatment with acetylshikonin ( Figure S6E-H). To further analyse haematopoietic toxicity, we counted the white blood cell (WBC) number after treatment with acetylshikonin. The results revealed that the WBC count was not affected by treatment with acetylshikonin (Figures 7d and S6D).
Tumour tissues were prepared for IHC analysis, and the expression of identified as a promising drug target (Hu et al., 2019;Kim et al., 2012;Sugimori et al., 2017). In this study, we have examined induction of autophagy, anti-oxidative, anti-inflammatory, anti-proliferative, anti-fertility, and anticancer effects (He, Li, Su, Huang, & Zhu, 2016;Pietrosiuk et al., 2006;Skrzypczak et al., 2015;Wu et al., 2011;Zeng, Zhu, & Su, 2018). Several studies have demonstrated that the compound inhibits growth of a variety of cancers, including colorectal, breast, liver, medullary thyroid carcinoma, pancreas, melanoma, and gastric cancers (Cho & Choi, 2015;Hasenoehrl et al., 2017;Kim, Lee, Park, & Choi, 2016;Kretschmer et al., 2012;Park et al., 2017;Vukic et al., 2017;Zeng, Liu, & Zhou, 2009). Mechanistically, acetylshikonin has been reported to act by downregulating NF-κB, Bcl-2 expression, and CYP2J2 activity, and the IL-8/MMP axis, or by up-regulating expression of haem oxygenase-1 (Cho et al., 2018;Cho & Choi, 2015). To identify a direct target for acetylshikonin, we examined the effect of acetylshikonin on the kinase activity of TOPK and MEK1 by an in vitro kinase assay ( Figure 1a,b). Acetylshikonin clearly inhibited TOPK kinase activity ( Figure 1a) but had no effect on MEK1 activity (Figure 1b). We used a kinase profiling assay to assess the effects of acetylshikonin on the kinase activity of several other kinases associated with TOPK signalling. The results indicated that acetylshikonin at 20 μM inhibited the kinase activity of Aurora A, Aurora B, and c-Src but had no effect on the activity of JNK1, ERK1, PDK1, or PKB ( Figure S1A). in vitro kinase assay further confirmed the inhibitory effects of acetylshikonin on Aurora A, Aurora B, and c-Src kinases. However, the compound had no effect on the catalytic activity of these kinases at or below 2.5-μM concentration ( Figure S1B-D). We have made prediction of binding scores between acetylshikonin and TOPK and compared that with the binding score of a reported TOPK inhibitor, 3-deoxysappanchalcone, by using computational modelling. The score between acetylshikonin to TOPK and that of 3-deoxysappanchalcone to TOPK is −6.94 and −5.02, respectively (Zhao et al., 2018). Moreover, the ATP competitive pull-down assay indicated that acetylshikonin might interact with the ATP-binding pocket of TOPK.
We also confirmed the ATP competitive binding of acetylshikonin with TOPK by kinetic analysis with an increasing K M value 15 to 20 (Figure 1e,f). Kim et al. (2012) reported the synthesis of a TOPK inhibitor, HI-TOPK-032, which also inhibited MEK1 activity by 40%, whereas acetylshikonin interacted with TOPK directly and inhibited TOPK kinase activity without affecting the activity of MEK1 The observed anti-proliferative effects of acetylshikonin was in accordance with the report of Zeng et al. (2009), who reported that acetylshikonin inhibited proliferation of PANC1 cells (Cho & Choi, 2015). Whereas acetylshikonin inhibited proliferation of PANC1 cells at 12.5 μM, it induced apoptosis and G1 phase arrest in HCT-15, HCT-116, SW 620, and DLD-1 colon cancer cell lines at 2.5 μM (Figure 3). This variation in effective concentration may be a cell type-specific phenomenon. However, it is notable that these colon cancer cell lines (HCT-15, HCT-116, SW 620, and DLD-1) express very high levels of TOPK ( Figure S2B). It is interesting to note that acetylshikonin had no effect on proliferation of HT-29 or SW 480 cells that express low levels of TOPK ( Figure S2C). Our results suggest that TOPK is a major target of acetylshikonin to inhibit proliferation and induce apoptosis in colon cancer cells. Several signalling molecules, including ERK, RSK, and c-Jun Zhu et al., 2007)  WBC count in an in vivo study (Matsuo et al., 2014;Nakamura et al., 2015). The present study revealed that acetylshikonin did not affect the WBC count in our mouse model (Figure 7d). It has been reported that some glioma stem cells are resistant to the TOPK inhibitor OTS964 (Sugimori et al., 2017). Whether acetylshikonin, as a TOPK inhibitor, can cause other serious side effects or can induce resistance in colon cancer cells needs to be examined in further investigations.
In summary, our study suggests that acetylshikonin can inhibit the growth of colorectal cancer tumours, in vitro and in vivo, by suppressing TOPK signalling through its direct targeting of TOPK, without modulation of MEK1 activity. As acetylshikonin inhibited tumour growth in the mouse model only when the PDX tumour tissue expressed high levels of TOPK, and showed no marked side effects, the compound or its derivatives can be considered for further clinical development of molecular target-based therapy for colorectal cancer.