Danshen improves survival of patients with colon cancer and dihydroisotanshinone I inhibit the proliferation of colon cancer cells via apoptosis and skp2 signaling pathway

Author’s Accepted Manuscript Danshen improves survival of patients with colon cancer and dihydroisotanshinone I inhibit the proliferation of colon cancer cells via apoptosis and skp2 signaling pathway Yin-Yin Lin, I-Yun Lee, Wen-Shih Huang, YuShin Lin, Feng-Che Kuan, Li-Hsin Shu, Yu-Ching Cheng, Yao-Hsu Yang, Ching-Yuan Wu

PII: DOI: Reference:

www.elsevier.com/locate/jep

S0378-8741(17)31772-5 http://dx.doi.org/10.1016/j.jep.2017.08.011 JEP10981

To appear in: Journal of Ethnopharmacology Received date: 4 May 2017 Revised date: 23 July 2017 Accepted date: 9 August 2017 Cite this article as: Yin-Yin Lin, I-Yun Lee, Wen-Shih Huang, Yu-Shin Lin, Feng-Che Kuan, Li-Hsin Shu, Yu-Ching Cheng, Yao-Hsu Yang and Ching-Yuan Wu, Danshen improves survival of patients with colon cancer and dihydroisotanshinone I inhibit the proliferation of colon cancer cells via apoptosis and skp2 signaling pathway, Journal of Ethnopharmacology, http://dx.doi.org/10.1016/j.jep.2017.08.011 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Danshen improves survival of patients with colon cancer and dihydroisotanshinone I inhibit the proliferation of colon cancer cells via apoptosis and skp2 signaling pathway

Yin-Yin Lin1#, I-Yun Lee1#, Wen-Shih Huang3,4, Yu-Shin Lin6 , Feng-Che, Kuan5, Li-Hsin Shu1,

Yu-Ching Cheng 1, Yao-Hsu Yang1,2, *, Ching-Yuan Wu 1,2,* 1

Department of Chinese Medicine, Chiayi Chang Gung Memorial Hospital, Chiayi, Taiwan.

2

School of Chinese medicine, College of Medicine, Chang Gung University, Tao-Yuan, Taiwan.

3

Division of Colon and Rectal Surgery, Department of Surgery, Chang Gung Memorial Hospital Chiayi,

Chiayi, Taiwan. 4

Chang Gung University College of Medicine, Taoyuan, Taiwan.

5

Department of Hematology and oncology, Chiayi Chang Gung Memorial Hospital, Chiayi, Taiwan.

6

Department of Pharmacy, Chiayi Chang Gung Memorial Hospital, Chiayi, Taiwan.

#

The two authors contributed equally to this work.

Correspondence Author:

Yao-Hsu Yang

Department of Chinese Medicine, Chiayi Chang Gung Memorial Hospital, Chiayi, Taiwan.

School of Chinese medicine, College of Medicine, Chang Gung University, Tao-Yuan, Taiwan.

No.6, W. Sec., Jiapu Rd., Puzi City, Chiayi County 613, Taiwan (R.O.C.). 1

Phone: +886 5 362-1000 Fax: +886 5 362-3002; E-mail: [email protected]

Ching Yuan Wu

Department of Chinese Medicine, Chiayi Chang Gung Memorial Hospital, Chiayi, Taiwan.

School of Chinese medicine, College of Medicine, Chang Gung University, Tao-Yuan, Taiwan.

No.6, W. Sec., Jiapu Rd., Puzi City, Chiayi County 613, Taiwan (R.O.C.).

Phone: +886 5 362-1000 Fax: +886 5 362-3002; E-mail: [email protected]

2

Abstract Ethnopharmacological relevance Danshen (Salvia miltiorrhiza Bunge) is widely used in traditional Chinese medicine. However, it’s definite clinical effect and mechanism on colon carcinoma is unclear. Aim of the study To test the hypothesis that the protective effect of danshen on colon cancer and discover the bioactive compounds through in vitro study. Materials and methods We conducted a nationwide cohort study by using population-based data from the Taiwan National Health Insurance Research Database (NHIRD). The study cohort comprised patients diagnosed with malignant neoplasm of colon (ICD-9-CM codes:153) in catastrophic illness database between January 1, 2000, and December 31, 2010. We used the Kaplan–Meier method to estimate lung cancer cumulative incidences. Next, human colon cancer cells (HCT 116 cells and HT29 cells) were used to investigate the effect of dihydroisotanshinone I (DT) on the proliferation and apoptosis of human colon cancer cells and the underlying mechanism through XTT assay and flow cytometry. The in vivo effect of DT treatment was investigated through a xenograft nude mouse model. Results

3

In our study, the in vivo protective effect of danshen in the different stage of colon cancer patients was validated through data from the National Health Insurance Research Database in Taiwan. In vitro, we found that dihydroisotanshinone I (DT), a bioactive compound present in danshen, can inhibit the proliferation of colon carcinoma cells, HCT 116 cells and HT-29 cells. Moreover, DT induced apoptosis of colorectal cancer cells. DT also repressed the protein expression of Skp2 (S-Phase Kinase Associated Protein 2) and the mRNA levels of its related gene, Snail1 (Zinc finger protein SNAI1) and RhoA (Ras homolog gene family, member A). In addition, DT also blocked the colon cancer cells recruitment ability of macrophage by decreasing CCL2 secretion in macrophages. DT treatment also significantly inhibited the final tumor volume in a xenograft nude mouse model. Conclusion Danshen has protective effects in colon cancer patients, which could be attributed to DT through blocking the proliferation of colon cancer cells through apoptosis.

Key words Dihydroisotanshinone I, colon carcinoma, Skp2, National Health Insurance Research Database

4

Abbreviations: NHIRD: National Health Insurance Research Database NHI: National Health Insurance CRC: Colorectal cancer Skp2: S-phase kinase-associated protein 2 CCL2: CC chemokine ligand 2 TAMs: tumor-associated macrophages TCM: Traditional Chinese Medicine FHPs: Finished herbal products EMT: tumor epithelial-mesenchymal transition Snail1: Zinc finger protein SNAI1 RhoA: Ras homolog gene family, member A DT: dihydroisotanshinone I TI: Tanshinone I T2A: Tanshinone IIA SA: Salvianolic acid B

5

Chemical compounds studied in this article: Dihydroisotanshinone I (PubChem CID:89406) Tanshinone I (PubChem CID: 114917) Tanshinone IIA (PubChem CID: 164676) Salvianolic acid B (PubChem CID: 6451084) Oxaliplatin (PubChem CID: 5310940)

6

Introduction Colorectal cancer (CRC) is the third most common malignant disease in worldwide (Schreuders et al., 2015). While resection is often curative, as many as 45% of CRC patients die as a result of the disease, despite treatment (Ferlay et al., 2010). In these advanced tumors show recurrences in distant organs such as the liver, lung, lymph node, bone or peritoneum even after complete resection of the primary tumors (Itatani et al., 2016). Therefore, developing treatment regimens with superior effectiveness and minimal adverse effects for CRC remain apriority in colorectal adenocarcinoma research. S-phase kinase-associated protein 2 (Skp2) belongs to the F-box protein family and is one of the components of the SCF E3 ubiquitin ligase complex. Skp2 overexpression has been observed in in human colorectal cancer and may have critical downstream effects in CRC development (Shapira et al., 2005; Tian et al., 2013). In addition, the previous reports showed that inhibition of skp2 could be a potential strategy for colorectal cancer (Bochis et al., 2015; Uddin et al., 2008). Chronic inflammation, including inflammatory bowel disease, plays an important contributing role to the development of colorectal cancer (Choi et al., 2016; Ekbom et al., 1990; Nowacki et al., 2015; Rogler, 2014; Yashiro, 2014). In the inflammation environment, tumor-associated

7

macrophages (TAMs) are derived from peripheral blood monocytes that are recruited into the tumor. TAMs also potentiate the seeding and establishment of metastatic cells(Guo et al., 2010). Chemokines are also one of the key players that promote cancer cell metastasis in CRC (Itatani et al., 2016). The expression of CC chemokine ligand 2 (CCL2, also known as monocyte chemoattractant protein-1) in CRC cells and TAM accumulation are strongly correlated with advanced CRC stages and a poor prognosis (Bailey et al., 2007; Hu et al., 2009). Moreover, the neutralizing antibody against CCL2 was reported to inhibit development of malignant pleural effusion of CRC in the animal model(Marazioti et al., 2013). These reports suggested CCL2 and skp2 could be the novel targets for the anti-CRC treatment. In Traditional Chinese Medicine (TCM), the dried root of Salvia miltiorrhiza Bunge (danshen) is used to treat numerous cardiovascular and endocrine diseases and cancers (Chen et al., 2001). However, the clinical effects and mechanism of danshen in colon cancer treatment remains unclear. The National Health Insurance Research Database (NHIRD) of Taiwan owned the almost complete information of patients in Taiwan, including the clinical drugs and TCM and is widely used to investigate the clinical effort of these drugs and TCM on patients in Taiwan(Chang et al., 2017; Hung et al., 2017; Liao et al., 2017; Lin et al., 2017; Liu et al., 2016). Finished herbal products

8

(FHPs), a modern form of decoctions in which herbal formulae and single herbs are concentrated into granulated compounds, are widely prescribed by TCM physicians because of their convenience and quality. The National Health Insurance program in Taiwan reimburses claims for FHPs. In this study, we used NHIRD to discover the clinical protective effect of danshen on colon cancer patients. In addition, we observed an inhibitory effect of dihydroisotanshinone I (DT) (Fig. 3A), extracted from the dried root of S. miltiorrhiza Bunge, on the proliferation of colon carcinoma cells, HCT 116 cells and HT-29 cells. DT also can induce apoptosis of HCT 116 cells and diminished the ability of macrophages to recruit HCT 116 cells. Mechanistically, DT could interrupt Skp2 and the tumor epithelial-mesenchymal transition (EMT) gene expression, including Snail1 (Zinc finger protein SNAI1) and RhoA (Ras homolog gene family, member A). DT also reduced the secretion of CCL2 from macrophages to inhibit the ability of recruiting colon tumor cells. Moreover, we found that DT treatment (30 mg/kg) significantly inhibited the final tumor volume on xenograft nude mice. Our result suggests that DT could be the novel candidate for anti-colorectal cancer in the further.

9

1. Material and methods 1.1. Data source We conducted a nationwide cohort study by using population-based data from the Taiwan National Health Insurance Research Database (NHIRD). Because National Health Insurance is a compulsory universal program for all residents in Taiwan, the NHIRD is a comprehensive health care database that covers nearly the entire 23.7 million populations of this country. We used databases for admissions and outpatient visits, both of which included information on patient characteristics such as sex, date of birth, date of admission, date of discharge, dates of visits, and up to five discharge diagnoses or three outpatient visit diagnoses (according to International Classification of Diseases, Ninth Revision (ICD-9) codes). The data files also contained information on patient prescriptions, including the names of prescribed drugs, dosage, duration, and total expenditure. Following strict confidentiality guidelines in accordance with personal electronic data protection regulations, the National Health Research Institutes of Taiwan maintains an anonymous database of NHI reimbursement data that is suitable for research. Meanwhile, this study was approved by the Ethics Review Board of Chang Gung Memorial Hospital, Chia-Yi Branch, Taiwan (201601433B1). 1.2. Study subjects

10

This study cohort was obtained from the Taiwanese National Health Insurance research database, which included all patients who received diagnosis of malignant neoplasm of colon (ICD-9-CM codes:153) in catastrophic illness database between January 1, 2000, and December 31, 2010. Patients who apply for a cancer catastrophic illness certificate are required to provide pathological reports or other supporting documents, such as laboratory and image studies. The date of the initial colon cancer diagnosis was defined as the index date of colon cancer. Patient with other cancer diagnosed before colon cancer or missing data were excluded. A total of 56,965 patients were included in the study cohort. Then these patients were categorized into three groups according to clinical cancer staging: stage I & II, stage III, stage IV (Fig. 1) depending on previous study (Kuan et al., 2017). These patients accrued follow-up time beginning on January 1, 2000 and ended on the date of death, or withdrawal from the registry or on December 31, 2010. 1.3. Danshen exposure and potential confounders and statistical analysis for NHIRD Finished herbal products (FHP) are the modern form of Chinese herbal remedies, of which single herb and herbal formulae are concentrated into granulated compounds, which fully reimbursed under the current NHI system of Taiwan. The list of reimbursed FHP was downloaded from the website of the Bureau of NHI. The corresponding drug

11

information for each FHP including the proportions of each constituent, date and period of approval as drug, code and name of manufacturer. By using this information, we determined the original amounts of danshen, in grams, for each mixture of FHPs. First, patients of each stage were categorized into 2 groups: had used danshen more than 84 grams after colon cancer diagnosed, and those with less than 84 grams danshen used in records. Moreover, patients of each stage also were categorized into 2 groups: had used danshen more than 28 days after colon cancer diagnosed, and those with less than 28 days with treatment of danshen in records. We used the Kaplan-Meier method to estimate survival probabilities and the log-rank test was performed to examine differences in the risk of death in the cohort. All of these analyses were conducted using SAS statistical software (Version 9.4; SAS Institute, Cary, NC, USA). 1.4. Cell culture and treatment The human colorectal cancer cell lines (HCT 116 cells, HT-29 cells) and mouse macrophage cell line (RAW264.7 cells) were obtained from the American Type Culture Collection. The RAW264.7 cells were cultured in Dulbecco's Modified Eagle's medium (Invitrogen Corp., Carlsbad, CA), supplemented with 10% FBS at 37℃ and 5% CO2. The HCT 116 cells and HT-29 cells were cultured in RPMI medium (Invitrogen Corp., Carlsbad, CA), supplemented with 10% fetal bovine serum at 37℃ and 5% CO2.

12

Dihydroisotanshinone I (DT) was obtained from ChemFaces Natural Products Co., Ltd., China (Catalog number: CFN-90162, the purity of dihydroisotanshinone I is 98% and its solubility in DMSO is >5mg/mL, PubChem CID:89406). Tanshinone I (TI) was obtained from was obtained from Sigma-Aldrich (Catalog number: SI-T5330, PubChem CID: 114917). Tanshinone IIA (T2A) was obtained from was obtained from Sigma-Aldrich (Catalog number: SI-T4952, PubChem CID: 164676). Salvianolic acid B (SA) was obtained from was obtained from Santa Cruz (Catalog number: sc-212911, PubChem CID: 6451084). Oxaliplatin was obtained from Sigma-Aldrich (Catalog number: SI-O9512, PubChem CID: 5310940). Human colorectal cancer cells and macrophages were cultured to 60-70% confluence prior to treatment. Medium was then replaced with fresh medium containing indicated compounds in DMSO (dimethyl sulfoxide) at the indicated concentrations. Cells treated with DMSO alone were used as untreated controls. 1.5. XTT assay The indicated colon cancer cell lines were plated at a density of 103 per well, in 96-well plates, in RPMI containing 10% FBS. Once attached, the medium was replaced with RPMI containing 10% FBS. The cells were then treated with indicated drugs for 24 or 48 hours; and absorbance were measured using the XTT assay kit (Roche, Cat. No.

13

11465015001) according to the manufacturer’s instructions as described previously (Wu et al., 2012). The XTT formazan complex was quantitatively measured at 492 nm using an ELISA reader (Bio-Rad Laboratories, Inc.). 1.6. Flow cytometry HCT 116 cells (1 × 106 cells) were seeded in a 100-mm plate and cultured overnight before treatment. Then, the cells were treated with control or 10 μM of indicated drugs for 24-48 hours. Then treated cells were detected by Annexin V-FITC Apoptosis Detection Kit (Strong Biotech Corporation, Cat No.: AVK250) and Mitoscreen JC-1 kit (BD Biosciences: 551302) according to the manufacturer’s instructions. In brief, at the end of the incubation period, the medium was removed. The treated cells were collected after washing by cold PBS. The supernatant was removed by centrifugation and then resuspended in indicated buffer by staining at room temperature in the dark for 15 min. The stained cells were analyzed by the flow cytometer BD FACSCanto (Becton Dickinson). Apoptosis of different developmental stages were studied by gating the respective population in the Dot Plots. 1.7. Cell Migration Assay Cell migration assays were performed as described previously (Lin et al., 2013). For the colorectal cancer recruitment assay, RAW264.7 cells (1 × 105 cells/well) were

14

treated with or without DMSO and with an indicated concentration of DT for 24 hours. The conditioned medium or control medium was then collected and plated in the lower chambers. The indicated parental human colorectal cancer cells (1 × 105 cells/well) were plated in the upper chambers in the medium without FBS. After incubation for 24 hours, the cells that have migrated into the bottom were fixed and stained using 1% toluidine blue, and the numbers were averaged after counting 6 randomly selected fields. Each sample was assayed in triplicate, and each experiment was repeated at least twice. 1.8. Quantitative real time PCR Total RNA was extracted from colorectal cancer cells using the TRIzol reagent (Invitrogen, Cat. No. 15596-026) according to the manufacturer’s instructions. Reverse transcription was performed using the Superscript first strand synthesis kit (Invitrogen, Number: 11904018). Quantitative real-time PCR analyses using the comparative CT method were performed on an ABI PRISM 7700 Sequence Detector System using the SYBR Green PCR Master Mix kit (Perkin Elmer, Applied Biosystems, Wellesley, MA, USA) according to the manufacturer’s instructions. Following initial incubation at 50°C for 2 minutes and 10 minutes at 95°C, amplification was performed for 40 cycles at 95°C for 20 seconds, 65°C for 20 seconds and 72°C for 30 seconds. Primers used were: Skp2 forward, 5'‐TTA GTC GGG AGA ACT TTC CAG GTG‐3' and Skp2 reverse, 5'

15

‐AGT CAC GTC TGG GTG CAG ATTT‐3'. RhoA forward, 5'‐GAG CAC ACA AGG CGG GAG‐3' and RhoA reverse, 5'‐CTT GCA GAG CAG CTC TCG TAG‐3'. Snail1 forward, 5'‐GAG GCG GTG GCA GAC TAG AGT‐3' and snail1 reverse, 5' ‐CGG GCC CCC AGA ATA GTTC‐3'. GAPDH forward, 5'‐TGC ACC ACC AAC TGC TTAGC‐3' and GAPDH reverse, 5'‐GGC ATG GAC TGT GGT CATGA‐3'. GAPDH was used as the housekeeping gene for data normalization. 1.9. Enzyme-linked immunosorbent assay (ELISA) The ELISA were performed as described previously (Izumi et al., 2013). Medium was collected from monoculture of macrophages under the treatment with or without DMSO or indicated concentration of DT for 24 hours. Mouse CCL2 in medium were detected by mouse CCL2 ELISA kits (eBioscience, catalog number: 88-7399) according to the manufacturer’s instructions. 1.10. Western blot analysis Western blot analyses were performed as described previously (Tsao et al., 2014). For Western blotting, cellular extracts of the human colorectal cancer cell line (HCT 116 cells) treated with DMSO or indicated concentrations of DT for 24 hours were prepared according to the manufacturer’s instructions. Equal amounts of protein were fractionated on a 10 % SDS-PAGE and transferred to polyvinylidene difluoride

16

membranes. The membranes were then blocked with 5% nonfat dried milk for 30 minutes and incubated in primary antibody for 6 hours at room temperature. The primary antibodies used were: anti-skp2 antibody (Cell Signaling, ratio: 1:1000), anti-β-actin antibody (Santa Cruz, IB: 1:10000). The primary antibodies and the secondary antibodies were diluted with 1% nonfat dried milk in 0.1% TBST (Tris-Buffered Saline Tween-20). Blots were washed by 0.1% TBST and incubated in horseradish peroxidase-conjugated secondary anti-mouse or anti-rabbit antibodies (Santa Cruz, ratio: 1:5000) for one hour at room temperature. After washing by 1X TBST again, protein signal was detected by chemiluminescence, using the Super Signal substrate (Pierce, Number: 34087). 1.11. Mouse Xenograft Model Mouse xenograft model were performed as described previously(Hu, Y. et al., 2015; Wu et al., 2017a). All procedures involving animals were approved by Animal Care and Use Committee (Approval number 2015060201) of Chang Gung Memory Hospital. Surgery was performed using sodium pentobarbital anesthesia. 10 male BALB/c-nu nude mice (18–20 g) aged 5–7 weeks were obtained from BioLASCO Taiwan Co., Ltd. and were used to build the model. HCT 116 cells were injected (1 × 106/Mouse) subcutaneously in the left and right flanks of nude mice. After 1 weeks, mice with

17

tumor sizes of about 15-20 mm3 were selected. Then, mice were randomized into 2 groups, 5 mice per each group and treated intraperitoneally with vehicle (2.5% DMSO) or with 30 mg/kg DT every 2 days. Tumor volume and mouse weight were measured every 2-3 days for 2 weeks. Tumor sizes were measured and tumor volume were calculated using the formula length x width x height x 0.52. Tumor size, body weight, and mortality of the mice were monitored daily. Following 2 weeks, the mice were sacrificed. 1.12.

Extraction Procedure of danshen extract

The dried root of Salvia miltiorrhiza Bunge (danshen) (100g, brought from Chang Gung Memory Hospital) were soaked in 1000 ml ethanol for 24 hours. The sample was filtered with filter paper while the residue was further extracted twice more under the same conditions. The filtrates collected from three separate extractions were combined and evaporated to dryness under vacuum. The dry extract was stored at -20°C. For all experiments, final concentrations of the tested compound were prepared by diluting the stock with ethanol. 1.13.

High-performance Liquid Chromatography (HPLC) conditions The HPLC analysis was performed with an SHIMADZU, LC-10Avp system.

Column: LiChrosper 100 RP-18e (4 mm i.d. x 250 mm, 5 μm), mobile phase: 0.05% trifluoroacetic acid-CH3CN (0 min, 95:5; 55 min, 0:100; 56 min, 95:5; 66 min, 95:5), Flow rate: 1.0 ml/min, wavelength: 280nm, Column temp: 40℃. 1.14. Statistical analyses All values were the means ± standard error of mean (SEM) of the replicate samples 18

(n=3 to 6, depending on the experiment) and experiments were repeated by a minimum of three times. Differences between two groups were assessed using the unpaired two-tailed Student’s t-test or by ANOVA if more than two groups were analyzed. The Tukey test was used as a post-hoc test in ANOVA for testing the significance of pairwise group comparisons. P-values <0.05 were considered statistically significant in all comparisons. SPSS version 13.0 for windows (LEAD technologies, Inc.) was used for all calculations. 2. RESULTS 2.1. Protective effect of danshen in colorectal cancer patients from Taiwan We examined 79833 patients diagnosed with prostate cancer within the years 2000-2010, and after excluding those who did not meet the study’s inclusion criteria, 56965 patients remained. In the initial data management of NHIRD data analysis, we had excluded missing data and error data. Besides, these databases have previously been used for epidemiologic research, and the information on prescription use, diagnoses, and hospitalizations is of high quality (Lai et al., 2010; Yang, Y.H. et al., 2015). Each pharmaceutical company in Taiwan has published and submitted the detailed composition of each of its products to the Committee on Chinese Medicine and Pharmacy to be approved for registration. By using this information, we determined the original amounts of danshen from Chinese herbal products during the exposure period. Then these patients were categorized into three groups according to clinical cancer staging: stage I & II, stage III, stage IV (Fig. 1). In each stage, patients were categorized into two groups according to drug dosage: those who had used either > 84 or ≤ 84 g 19

of danshen after their colon cancer diagnosis, as per medical records. Patients were also categorized into two groups according to the duration of their drug use: those who had used danshen for either > 28 or ≤ 28 days after their colon cancer diagnosis, as per medical records. Because danshen was common used herbs for colon cancer and the number of danshen non-user is too small, the number of danshen non-users were included into the groups of

≤ 84 g or ≤ 28 days. After 10 years of follow-up, the

survival rate analyses demonstrated a strong dose-dependent and time-dependent association between the use of danshen and survival (Fig. 2). Notably, the patients who used > 84 g of danshen or danshen used for > 28 days exhibited an increase of about 5% -20% in the survival rate compared with the patients who used ≤ 84 g of danshen or danshen used for ≤ 28 days in different stage of colon cancer patients. Thus, these data demonstrated the protective effects for higher dose or longer used of danshen for colon cancer patients in Taiwan. 2.2. The effect of dihydroisotanshinone (DT) on the growth of colon cancer cells From NHIRD, we discovered the protect effect of danshen for different stage of the colon cancer patients in Taiwan. Depending on their structure and properties, there are at least more than 50 compounds isolated from danshen (Wang et al., 2007). These compounds extracted from danshen have been divided into two groups. The first group is phenolic acids, such as salvianolic acid B (SA), and their structure contain caffeic acid monomers and oligomers. The second group is tanshinones, such as tanshinone I (TI), tanshinone IIA (IIA) and dihydroisotanshinone I (DT), and their structure contain abietane diterpenes with a common ortho- or para-naphthoquinone chromophore (Fig. 3A). To study the effect of these compounds on proliferation of colon cancer cells, we used the human colorectal carcinoma cell line (HCT 116 cells) and human colorectal 20

adenocarcinoma cell line (HT-29 cells) as our model to investigate the effect of these compounds, including SA, TI, IIA and DT, in the XTT assay (Fig. 3B, C). After treatment with indicated compounds for 24 hours, our results showed that tanshinones, including TI, IIA and DT, significantly inhibited proliferation of HCT 116 cells in a dose-dependent manner for 24 hours (Fig. 3B). Next, we also discover that the 20-30 μM DT had the inhibitory effect on the proliferation of HT-29 cells (Fig. 3C). In the previous study, IIA induces less intracellular ROS and cytotoxicity effect in HT29 cells than those in HCT116 cells (Liu et al., 2013). The effect of DT on these colon cancer cell lines is similar to this report (Liu et al., 2013). Notably, that 5 μM tanshinones, including IIA and DT, exhibited the better inhibitory effect compared with 5 μM oxaliplatin (OXA), the clinical common used anti-colon cancer agent. Moreover, DT had the best inhibitory effect on proliferation of HCT 116 cells even in the lower concentration (5 μM). These data suggested the role of DT in growth-inhibition of human colon cancer cells. In addition, other tanshinones, like IIA, also contribute to the anti-colon cancer effect. 2.3. The effects of DT on apoptosis of HCT 116 cells in vitro For further study whether the inhibition of cell proliferation induced by these pure compounds extracted from danshen were associated with apoptosis, we determined the mechanism of these compounds in HCT 116 cells. Cells were treated with indicated compounds for 24-48 hours, and then the cells were analyzed for apoptosis by flow cytometry with annexin V/PI dual staining. The results demonstrated that tanshinones, including TI, IIA and DT, significantly induced apoptosis of HCT 116 cells in a time-dependent and dose-dependent manner (Fig. 4A). As the previous result, DT had the best induced-apoptosis effect (about 33.7%-56.8%) on HCT 116 cells even in the 21

lower concentration (5 μM). Moreover, we also confirm HCT 116 cells treated with DT were undergoing apoptosis through mitochondrial depolarization by using JC-1 staining assay (Fig. 4B). Gao et al reported that total tanshinones showed more cytotoxic effects compared with tanshinone IIA through apoptosis evidences, including annexin V/7-AAD double staining, the depolarization of mitochondrial membrane potential and the up-regulation of pro-apoptotic proteins on lung cancer cells (Gao et al., 2015). These results suggest that apoptosis is the important mode of cell death induced by these tanshinones in colon cancer cells. DT has better effect to inducing apoptosis compared with IIA or TI in HCT 116 cells. 2.4. DT decrease Skp2 protein expression and mRNA level of the related EMT genes and inhibits the colon cancer cells recruitment ability of macrophages through decreasing the CCL2 protein secretion from macrophages In previous reports, skp2 can enhance tumor metastasis through modulating the proteins of EMT, including MMP-9 and snail (Hung et al., 2010; Lu et al., 2014; Wei et al., 2013; Yang et al., 2014). The RhoA (Ras homolog gene family, member A) GTPase is crucial for cancer metastasis and RhoA transcription is regulated by skp2 complex (Chan et al., 2010). To determine how DT regulates the growth of colon cancer cells, we also determined the protein expression of Skp2 through Western blot analysis in HCT 116 cells treated with DMSO or 10 μM DT. Our results revealed that DT inhibited the protein expression of Skp2 in HCT 116 cells (Fig. 5A). We also determined the mRNA expression of skp2, snail1 and RhoA through qPCR. As shown in Fig. 5B-D, DT

22

inhibited the mRNA expression of skp2, snail1 and RhoA in HCT 116 cells. Our data suggested that DT inhibits the growth in HCT 116 cells by modulating the protein expression of Skp2 and the mRNA level of the down-stream genes. Previous studies have shown that macrophages is able to promote tumor invasion and metastasis(Huang et al., 2015), and that the human prostate cancer cells recruitment’s ability of macrophages was inhibited by DT(Wu et al., 2017b), therefore we investigated the effect of DT on the ability of macrophages to promote colon cancer cells migration. We used the mouse macrophage-like cell line, RAW 264.7 cells, in our model [19]. After RAW 264.7 cells were treated with or without DMSO and with 5 or 10 μM DT for 24 hours, the conditioned medium was collected and then placed in the lower chambers of the transwell plates. The HCT 116 cells were then placed in the upper chambers of the transwell plates with inserts in serum-free medium for the migration assay (Fig. 5E). Our results indicated that 5–10 μM DT inhibited the migration of HCT 116 cells in a dose-dependent manner in the RAW 264.7 cell medium (Fig. 5F). The expression of CCL2 in CRC cells and TAM accumulation are strongly correlated with advanced CRC stages and a poor prognosis (Bailey et al., 2007; Hu et al., 2009). CCL2 also promote metastasis of several cancers, including nasopharyngeal carcinoma, bladder cancer, thyroid carcinoma and colon cancer (Passaro et al., 2016; Rao et al., 2016; Volpato et al., 2016; Yang, J. et al., 2015). A recent study found that CCL2 creates an environment for bone metastasis by increasing the infiltration of TAMs (Craig and Loberg, 2006). In our previous data, the ability of macrophages to promote the migration of colorectal cancer cells was inhibited by DT. We subsequently examined the effect of DT on the secretion of CCL2 from macrophages using ELISA. After treatment with or without the indicated drugs and vehicle control (DMSO) for 24 hours, 23

we found that DT treatment inhibited the secretion of CCL2 from RAW 264.7 cells in a dose dependent manner (Fig. 5G). Our data suggested that DT may inhibit the migration of colorectal cancer cells in macrophage’s medium through its effects on CCL2. 2.5. The in vivo effect of DT on a xenograft nude mouse model For an in vivo study, we investigated the effects of DT on a xenograft nude mouse model. DT treatment (30 mg/kg, IP) significantly inhibited the final tumor volume by approximately 80% in 2 weeks (Fig. 6B), whereas DT did not significantly alter either the activity or body weight of mice, and no mice were dead after DT treatment (30 mg/kg, IP) for 2 weeks (Fig. 6A). These results suggested that DT treatment exerts limited adverse effects on mice, thus validating our data from cell lines. 3. Discussion To the best of our knowledge, this is the first nationwide-cohort population study that analyzed the clinical effects of danshen on the survival rate of colon cancer patients. Specifically, we examined data from the computerized insurance reimbursement claims database in Taiwan, and we used the NHIRD to discover that higher dose or longer used of danshen can prolong the survival rate of colon cancer patients in different stages. Notably, the patients with stage I & II and IV colon cancer exhibited the better survival rate (about 20 %) compared with the patients with stage III colon cancer (about 5%) after danshen used. It is possible that the smaller size of the stage 3 colon cancer patients (N=4563) compared with the group with stage I & II (N=28843) and stage IV

24

(N=23559). Moreover, there were some limitations with this study that should be noted. The danshen decoctions can be obtained from pharmacies in Taiwan. In our study, the medical record from the NHIRD offered reimbursements only for FHPs that were prescribed by TCM physicians, and not for decoctions provided by pharmacies; this may have led to an underestimation of the TCM utilization dosage. However, this underestimation could be small because decoctions was expensive and not common in Taiwan. FHPs were reimbursed by NHI and most common used by clinical TCM doctors in Taiwan. In clinical Chinese medicine, danshen was usually used with the formulation mixed with other herbs. It can’t exclude the protective effect of other herbs on the colon cancer patients from the result of NHIRD. In our NHIRD result, patients were categorized into these groups according to drug dosage: those who had used either > 84 or ≤ 84 g of danshen after their colon cancer diagnosis or either > 28 or ≤ 28 days after their colon cancer diagnosis, as per medical records. For the two groups of those who had used > 84 g and > 28 days of danshen after their colon cancer diagnosis, danshen could be major compound in their formulation. The effect of other herbs in these groups should be less than danshen. However, although we determined that the use of danshen could prolong the survival rate of colon cancer patients in Taiwan, more rigorous, randomized, double-blind, and placebo-controlled trials are necessary to

25

confirm the protective effect of danshen in colon cancer patients in the future. Abietane diterpenes that have a common ortho- or para-naphthoquinone chromophore are the major components of tanshinones and include tanshinone I, tanshinone IIA, cryptotanshinone and DT (Wang et al., 2007). Kong et al reported that the concentration of DT in danshen injection is about 0.0125 % (Kong and Liu, 1984). In our HPLC result, the concentration of DT in danshen extract is about 0.025% (Fig. 7). These results suggest that the different protocols produce the different concentrations of DT from danshen. In Taiwan, the different pharmaceutical companies have different protocols to produce their finished herbal products. It is possible these concentrations of these active materials are different in these finished herbal products from different pharmaceutical companies. Moreover, the National Health Insurance program in Taiwan reimburses claims for these finished herbal products. For this reason, we could not verify the exact dosage of these active materials that the study participants ingested. To confirm the exact effect of these active materials, we performed the in vitro studies to validate the result from NHIRD. In previous studies, several members of tanshinones, including TI, IIA or dihydrotanshinone I, can induce apoptosis or autophagy in colon cancer (Lu et al., 2016; Su et al., 2008a; Yoshimura et al., 2015). IIA and TI can inhibit the growth and induced apoptosis in different human colon cancer cells(Bai et al., 2016;

26

Hu, T. et al., 2015; Liu et al., 2013; Lu et al., 2016; Su, 2012; Su et al., 2008a; Su et al., 2008b; Sui et al., 2014; Wang et al., 2014). Our result about the inhibitory effect of IIA and TI on colon cancer cells lines are consistent with these previous studies. For dihydrotanshinone which is isomer of DT, the inhibitory effect on the growth of colon cancer cells was also reported(Hu et al., 2014; Wang et al., 2015). In our result, DT have significantly inhibitory effect on the proliferation on colon cancer. Notedly, the inhibitory effect of DT is better than IIA and TI, even better than oxaliplatin. Moreover, mitochondria mediated caspase dependent pathway was major pathway in cytotoxicity of dihydrotanshinone (Wang et al., 2015). Our result showed DT also induces apoptosis through mitochondrial depolarization (Fig. 4B). This result is consistent with the previous report (Wang et al., 2015). Moreover, the inducing-apoptosis effect of DT is better than IIA and TI in our data (Fig. 4A). IIA and DT possess an ortho-quinone and an intact ring D (Fig. 3A), respectively, suggesting that the structure of tanshinones may influence the ways in which they inhibit the proliferation of colon cancer cells. DT treatment also significantly inhibited the final tumor volume in a xenograft nude mouse model. Taken together, these results suggest that DT could be considered as the better candidate for colon cancer therapy compared with IIA and TI. Kang et al demonstrated that intra-tumoural TAM increased colorectal cancer cell

27

invasiveness and migration and count correlated with parameters of worse disease progression (Kang et al., 2010). In previous study, Shan et al (Shan et al., 2009) showed that IIA can block invasion and metastasis of the other different CRC cell lines, HT29 and SW480 cells. IIA also can inhibit the proliferation of HCT116 and HT-29 cells by reducing the production of inflammatory cytokines tumor necrosis factor α and interleukin 6 from RAW264.7 cells through repression of microRNA-155 (Tu et al., 2012). However, the effect of DT on the mobility ability and cross talk between CRC and macrophage is still undiscovered. Our novel study demonstrate that DT is able to inhibit the migration promoting ability of macrophages in colorectal cancer. It suggests that DT may affect the cross talk between macrophages and tumors. It has been shown that IIA exerted cardioprotective effects by reducing CCL2 secretion by cardiac fibroblasts (Ren et al., 2010). Our results demonstrated that 5–10 μM DT is able to reduce the secretion of CCL2 from RAW264.7 cells. It suggests that CCL2 is the major inhibitory target from macrophage under the treatment of DT. IIA can inhibit hypoxia-induced pulmonary artery smooth muscle cell proliferation via the Skp2 pathway (Luo et al., 2013). For colorectal cancer, several studies showed that skp2 play a critical role during the development and metastasis of cancer (Li et al., 2004; Timmerbeul et al., 2006). Skp2 can enhance tumor metastasis through modulating

28

the proteins of EMT, including MMP-9 and snail1 (Hung et al., 2010; Lu et al., 2014; Wei et al., 2013; Yang et al., 2014). RhoA transcription is regulated by skp2 complex and is crucial for cancer metastasis (Chan et al., 2010). For colorectal cancer, decreased activation of RhoA and Rac1 can increase cell–cell contact and decreased actin cytoskeletal remodeling to inhibit the CRC metastasis(Gulhati et al., 2011). In our study, we discovered that DT reduces the protein expression of Skp2 and down-regulated the mRNA levels of snail1 and RhoA. It suggests that skp2 pathway may play an important role in the effect of DT on colorectal cancer. Oxaliplatin-based chemotherapy is a standard of care agent used most commonly in combination with 5-fluorouracil in patients with colorectal cancer. Unfortunately, resistance to oxaliplatin based combinations is the major reason of treatment failure (Martinez-Balibrea et al., 2015). In our study, 5 μM DT showed the better inhibitory effect on proliferation of colorectal cancer cells than the same concentration of oxaliplatin (Fig. 3B). It suggests DT could be the more potential candidate of anti-colorectal cancer compound than oxaliplatin. In summary, our data suggest that DT is capable of inhibiting proliferation of colorectal cancer cells through the regulatory effects via the Skp2 pathway. DT also has the ability to interrupt the cross talk between colorectal cancer cells and macrophages through CCL2. These results suggest that DT might be a novel anticancer agent in the

29

armamentarium of colorectal cancer management.

30

Disclosure of potential conflicts of interest The authors declare no conflict of interest.

Acknowledgments This work was supported by grant CMRPG6C0262, CMRPG6D0343, CMRPG6F0011, CMRPG6F0661, CMRPG6E0091, CMRPG6F0551 and CMRPG6F0171 from Chang Gung

Memorial

Hospital,

and

MOST

105-2320-B-182-006-MY3,

NSC101-2320-B-182A-005 and NSC 102-2320-B-182A-007 from the Ministry of Science and Technology.

Authors’ contributions YYL and IYL contributed to the conception and design of the entire study. YSL and YYL carried out most of the experiments, contributed to the data interpretation and wrote the initial draft of the manuscript. YHY contributed to the experimental design. YSL and WSH contributed to data interpretation and the final editing of the manuscript. IYL, YCC, FCK and LHS contributed to Western’s blotting assay, migration assay and cell cycle assay. CYW and YSL provided grant support, funding and the final editing of the manuscript. All authors read and approved the final manuscript.

31

Legends to figures Fig. 1. Patient of colon cancer disposition Fig. 2. The effect of danshen on the survival rate of Taiwan colon cancer patients. A total of 56,965 prostate cancer patients were included in the study cohort. These patients accrued follow-up time for 10 years. Crude overall Kaplan-Meier survival curves for the colon cancer patients was investigated. (A, C, E) The patients of each stage were categorized into 2 groups: never used danshen, had used danshen more than 84 grams after prostate cancer diagnosed, and those with less than 84 grams danshen used in records. (log-rank: p<0.001). (B, D, F) The patients of each stage were categorized into 2 groups: had used danshen more than 28 days after prostate cancer diagnosed, and those with less than 28 days with treatment of danshen in records (log-rank: p<0.001). Fig. 3. DT block the proliferation of colon cancer cell lines. (A) The structure of dihydroisotanshinone I (DT), tanshinone IIA, tanshinone I and salvianolic acid B. (B, C) HCT 116 cells or HT-29 cells were measured by XTT assay after indicated hours of culturing in the presence of indicated compounds. All the results are representative of at least three independent experiments. (Error bars=mean±S.E.M. Asterisks (*) mark samples significantly different from DMSO group with p<0.05).

32

Fig. 4. DT induces apoptosis in colon cancer cells. HCT 116 cells were treated without or with indicated compounds for 24-48h. Cell apoptosis was detected by flow cytometry with annexin-V-FITC/PI dual staining or mitoscreen JC-1 staining. (A) For annexin-V-FITC/PI dual staining, the representative histograms of flow cytometric analysis using double staining with annexin-V-FITC (FITC‑ A) and PI (PI‑ A). Q1 (annexin-V−/PI+) show necrosis cells; Q2 (annexin-V+/PI+) show the late apoptosis cells; Q3 (annexin-V−/PI−) show normal cells; Q4 (annexin-V+/PI−) show the early apoptosis cells. (B) For mitoscreen JC-1 staining, dot Plots revealing depolarization of mitochondria in treated HCT 116 cells. The percentage of events in the upper gate (P2) and lower gate (P3) represent population of treated HCT 116 cells having normal and depolarized mitochondria respectively. Fig. 5. DT inhibits the protein expression of Skp2 and the mRNA level of related EMT genes and the colorectal cancer cells recruitment of RAW264.7 in vitro. Total cell extracts of HCT 116 cell (A) were harvested from untreated cells and cells treated with indicated drugs for 24 hours. The protein was immunoblotted with polyclonal antibodies specific for skp2. β-actin was used as an internal loading control. Total mRNA was extracted from the HCT 116 cells after treat with indicated drugs for 24 hours. The coding regions of human skp2 (B), RhoA (C) and snail1 (D) were used as probes for real

33

time polymerase chain reaction analysis. The migration ability of HCT 116 cells in the RAW 264.7 cells medium were measured by the transwell migration assay. The RAW 264.7 cells were treated with indicated drugs for 24 hours. Then the condition medium were collected and placed in the lower chamber. The HCT 116 cells were then placed on the upper chamber for the migration assay(E). After incubation for 24 hours, the photographs (× 100) were taken and the migratory cells were measured using AlphaEase®FC StandAlone Software. Numbers of the migratory HCT 116 cells in each group were normalized to the control (F). For secretion of CCL2 from RAW264.7 cells. The condition medium of RAW264.7 cells were collected from untreated cells and cells treated with DMSO or indicated drugs for 24 hours. The secretion of mouse CCL2 were measured by ELISA kits (G). All the results are representative of at least three independent experiments. (Error bars=mean±S.E.M. Asterisks (*) mark samples significantly different from blank group with p<0.05). Fig. 6. The in vivo effect of DT on xenografted animal model. (A) Average mice weights with every 2-day injection of vehicle/DT over a time course of 3 weeks. (B) Average tumor volume of mice injected with either vehicle (DMSO) or DT (30 mg/kg, n = 5 per group). (Error bars=mean±S.E.M.) Fig. 7. HPLC chromatograms of dihroisotanshinone I (A), tanshinone IIA (B) and danshen extract (C). The concentration of dihydroisotanshinone I and tanshinone IIA in 34

danshen extract(D).

35

REFERENCES Bai, Y., Zhang, L., Fang, X., Yang, Y., 2016. Tanshinone IIA enhances chemosensitivity of colon cancer cells by suppressing nuclear factor-kappaB. Exp Ther Med 11(3), 1085-1089. Bailey, C., Negus, R., Morris, A., Ziprin, P., Goldin, R., Allavena, P., Peck, D., Darzi, A., 2007. Chemokine expression is associated with the accumulation of tumour associated macrophages (TAMs) and progression in human colorectal cancer. Clin Exp Metastasis 24(2), 121-130. Bochis, O.V., Irimie, A., Pichler, M., Berindan-Neagoe, I., 2015. The role of Skp2 and its substrate CDKN1B (p27) in colorectal cancer. J Gastrointestin Liver Dis 24(2), 225-234. Chan, C.H., Lee, S.W., Li, C.F., Wang, J., Yang, W.L., Wu, C.Y., Wu, J., Nakayama, K.I., Kang, H.Y., Huang, H.Y., Hung, M.C., Pandolfi, P.P., Lin, H.K., 2010. Deciphering the transcriptional complex critical for RhoA gene expression and cancer metastasis. Nat Cell Biol 12(5), 457-467. Chang, C.M., Wu, P.C., Chiang, J.H., Wei, Y.H., Chen, F.P., Chen, T.J., Pan, T.L., Yen, H.R., Chang, H.H., 2017. Integrative therapy decreases the risk of lupus nephritis in patients with systemic lupus erythematosus: A population-based retrospective cohort study. Journal of ethnopharmacology 196, 201-212. Chen, Y.L., Yang, S.P., Shiao, M.S., Chen, J.W., Lin, S.J., 2001. Salvia miltiorrhiza inhibits intimal hyperplasia and monocyte chemotactic protein-1 expression after balloon injury in cholesterol-fed rabbits. Journal of cellular biochemistry 83(3), 484-493. Choi, J.K., Kim, D.W., Shin, S.Y., Park, E.C., Kang, J.G., 2016. Effect of Ulcerative Colitis on Incidence of Colorectal Cancer: Results from the Nationwide Population-Based Cohort Study (2003-2013). J Cancer 7(6), 681-686. Craig, M.J., Loberg, R.D., 2006. CCL2 (Monocyte Chemoattractant Protein-1) in cancer bone metastases. Cancer metastasis reviews 25(4), 611-619. Ekbom, A., Helmick, C., Zack, M., Adami, H.O., 1990. Ulcerative colitis and colorectal cancer. A population-based study. N Engl J Med 323(18), 1228-1233. Ferlay, J., Shin, H.R., Bray, F., Forman, D., Mathers, C., Parkin, D.M., 2010. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. International journal of cancer. Journal international du cancer 127(12), 2893-2917. Gao, H., Sun, W., Zhao, W., Hao, W., Leung, C.H., Lu, J., Chen, X., 2015. Total Tanshinones-Induced Apoptosis and Autophagy Via Reactive Oxygen Species in Lung Cancer 95D Cells. The American journal of Chinese medicine 43(6), 1265-1279. 36

Gulhati, P., Bowen, K.A., Liu, J., Stevens, P.D., Rychahou, P.G., Chen, M., Lee, E.Y., Weiss, H.L., O'Connor, K.L., Gao, T., Evers, B.M., 2011. mTORC1 and mTORC2 regulate EMT, motility, and metastasis of colorectal cancer via RhoA and Rac1 signaling pathways. Cancer research 71(9), 3246-3256. Guo, R., Wang, T., Shen, H., Ge, H.M., Sun, J., Huang, Z.H., Shu, Y.Q., 2010. Involvement of mTOR and survivin inhibition in tamoxifen-induced apoptosis in human hepatoblastoma cell line HepG2. Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie 64(4), 249-253. Hu, H., Sun, L., Guo, C., Liu, Q., Zhou, Z., Peng, L., Pan, J., Yu, L., Lou, J., Yang, Z., Zhao, P., Ran, Y., 2009. Tumor cell-microenvironment interaction models coupled with clinical validation reveal CCL2 and SNCG as two predictors of colorectal cancer hepatic metastasis. Clinical cancer research : an official journal of the American Association for Cancer Research 15(17), 5485-5493. Hu, T., To, K.K., Wang, L., Zhang, L., Lu, L., Shen, J., Chan, R.L., Li, M., Yeung, J.H., Cho, C.H., 2014. Reversal of P-glycoprotein (P-gp) mediated multidrug resistance in colon cancer cells by cryptotanshinone and dihydrotanshinone of Salvia miltiorrhiza. Phytomedicine : international journal of phytotherapy and phytopharmacology 21(11), 1264-1272. Hu, T., Wang, L., Zhang, L., Lu, L., Shen, J., Chan, R.L., Li, M., Wu, W.K., To, K.K., Cho, C.H., 2015. Sensitivity of apoptosis-resistant colon cancer cells to tanshinones is mediated by autophagic cell death and p53-independent cytotoxicity. Phytomedicine : international journal of phytotherapy and phytopharmacology 22(5), 536-544. Hu, Y., Zhou, J., Ye, F., Xiong, H., Peng, L., Zheng, Z., Xu, F., Cui, M., Wei, C., Wang, X., Wang, Z., Zhu, H., Lee, P., Zhou, M., Jiang, B., Zhang, D.Y., 2015. BRD4 inhibitor inhibits colorectal cancer growth and metastasis. International journal of molecular sciences 16(1), 1928-1948. Huang, Z., Wu, T., Liu, A.Y., Ouyang, G., 2015. Differentiation and transdifferentiation potentials of cancer stem cells. Oncotarget 6(37), 39550-39563. Hung, K.F., Hsu, C.P., Chiang, J.H., Lin, H.J., Kuo, Y.T., Sun, M.F., Yen, H.R., 2017. Complementary Chinese herbal medicine therapy improves survival of patients with gastric cancer in Taiwan: A nationwide retrospective matched-cohort study. Journal of ethnopharmacology 199, 168-174. Hung, W.C., Tseng, W.L., Shiea, J., Chang, H.C., 2010. Skp2 overexpression increases the expression of MMP-2 and MMP-9 and invasion of lung cancer cells. Cancer Lett 288(2), 156-161. Itatani, Y., Kawada, K., Inamoto, S., Yamamoto, T., Ogawa, R., Taketo, M.M., Sakai, Y., 2016. The Role of Chemokines in Promoting Colorectal Cancer Invasion/Metastasis. 37

International journal of molecular sciences 17(5). Izumi, K., Fang, L.Y., Mizokami, A., Namiki, M., Li, L., Lin, W.J., Chang, C., 2013. Targeting the androgen receptor with siRNA promotes prostate cancer metastasis through enhanced macrophage recruitment via CCL2/CCR2-induced STAT3 activation. EMBO Mol Med 5(9), 1383-1401. Kang, J.C., Chen, J.S., Lee, C.H., Chang, J.J., Shieh, Y.S., 2010. Intratumoral macrophage counts correlate with tumor progression in colorectal cancer. J Surg Oncol 102(3), 242-248. Kong, D.Y., Liu, X.J., 1984. [Structure of dihydroisotanshinone I of danshen]. Yao xue xue bao = Acta pharmaceutica Sinica 19(10), 755-759. Kuan, F.C., Lai, C.H., Ku, H.Y., Wu, C.F., Hsieh, M.C., Liu, T.W., Yeh, C.Y., Lee, K.D., 2017. The survival impact of delayed surgery and adjuvant chemotherapy on stage II/III rectal cancer with pathological complete response after neoadjuvant chemoradiation. International journal of cancer. Journal international du cancer 140(7), 1662-1669. Lai, M.N., Wang, S.M., Chen, P.C., Chen, Y.Y., Wang, J.D., 2010. Population-based case-control study of Chinese herbal products containing aristolochic acid and urinary tract cancer risk. J Natl Cancer Inst 102(3), 179-186. Li, J.Q., Wu, F., Masaki, T., Kubo, A., Fujita, J., Dixon, D.A., Beauchamp, R.D., Ishida, T., Kuriyama, S., Imaida, K., 2004. Correlation of Skp2 with carcinogenesis, invasion, metastasis, and prognosis in colorectal tumors. International journal of oncology 25(1), 87-95. Liao, H.H., Yen, H.R., Muo, C.H., Lee, Y.C., Wu, M.Y., Chou, L.W., Sun, M.F., Chang, T.T., 2017. Complementary traditional Chinese medicine use in Children with cerebral palsy: a nationwide retrospective cohort study in Taiwan. BMC Complement Altern Med 17(1), 155. Lin, T.H., Izumi, K., Lee, S.O., Lin, W.J., Yeh, S., Chang, C., 2013. Anti-androgen receptor ASC-J9 versus anti-androgens MDV3100 (Enzalutamide) or Casodex (Bicalutamide) leads to opposite effects on prostate cancer metastasis via differential modulation of macrophage infiltration and STAT3-CCL2 signaling. Cell Death Dis 4, e764. Lin, T.H., Yen, H.R., Chiang, J.H., Sun, M.F., Chang, H.H., Huang, S.T., 2017. The use of Chinese herbal medicine as an adjuvant therapy to reduce incidence of chronic hepatitis in colon cancer patients: A Taiwanese population-based cohort study. Journal of ethnopharmacology. Liu, C.Y., Chu, J.Y., Chiang, J.H., Yen, H.R., Hsu, C.H., 2016. Utilization and prescription patterns of traditional Chinese medicine for patients with hepatitis C in Taiwan: a population-based study. BMC Complement Altern Med 16(1), 397. 38

Liu, M., Wang, Q., Liu, F., Cheng, X., Wu, X., Wang, H., Wu, M., Ma, Y., Wang, G., Hao, H., 2013. UDP-glucuronosyltransferase 1A compromises intracellular accumulation and anti-cancer effect of tanshinone IIA in human colon cancer cells. PloS one 8(11), e79172. Lu, H., Cao, X., Zhang, H., Sun, G., Fan, G., Chen, L., Wang, S., 2014. Imbalance between MMP-2, 9 and TIMP-1 promote the invasion and metastasis of renal cell carcinoma via SKP2 signaling pathways. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine 35(10), 9807-9813. Lu, M., Wang, C., Wang, J., 2016. Tanshinone I induces human colorectal cancer cell apoptosis: The potential roles of Aurora A-p53 and survivin-mediated signaling pathways. International journal of oncology 49(2), 603-610. Luo, Y., Xu, D.Q., Dong, H.Y., Zhang, B., Liu, Y., Niu, W., Dong, M.Q., Li, Z.C., 2013. Tanshinone IIA inhibits hypoxia-induced pulmonary artery smooth muscle cell proliferation via Akt/Skp2/p27-associated pathway. PloS one 8(2), e56774. Marazioti, A., Kairi, C.A., Spella, M., Giannou, A.D., Magkouta, S., Giopanou, I., Papaleonidopoulos, V., Kalomenidis, I., Snyder, L.A., Kardamakis, D., Stathopoulos, G.T., 2013. Beneficial impact of CCL2 and CCL12 neutralization on experimental malignant pleural effusion. PloS one 8(8), e71207. Martinez-Balibrea, E., Martinez-Cardus, A., Gines, A., Ruiz de Porras, V., Moutinho, C., Layos, L., Manzano, J.L., Buges, C., Bystrup, S., Esteller, M., Abad, A., 2015. Tumor-Related Molecular Mechanisms of Oxaliplatin Resistance. Mol Cancer Ther 14(8), 1767-1776. Nowacki, T.M., Bruckner, M., Eveslage, M., Tepasse, P., Pott, F., Thoennissen, N.H., Hengst, K., Ross, M., Bettenworth, D., 2015. The risk of colorectal cancer in patients with ulcerative colitis. Dig Dis Sci 60(2), 492-501. Passaro, C., Borriello, F., Vastolo, V., Di Somma, S., Scamardella, E., Gigantino, V., Franco, R., Marone, G., Portella, G., 2016. The oncolytic virus dl922-947 reduces IL-8/CXCL8 and MCP-1/CCL2 expression and impairs angiogenesis and macrophage infiltration in anaplastic thyroid carcinoma. Oncotarget 7(2), 1500-1515. Rao, Q., Chen, Y., Yeh, C.R., Ding, J., Li, L., Chang, C., Yeh, S., 2016. Recruited mast cells in the tumor microenvironment enhance bladder cancer metastasis via modulation of ERbeta/CCL2/CCR2 EMT/MMP9 signals. Oncotarget 7(7), 7842-7855. Ren, Z.H., Tong, Y.H., Xu, W., Ma, J., Chen, Y., 2010. Tanshinone II A attenuates inflammatory responses of rats with myocardial infarction by reducing MCP-1 expression. Phytomedicine : international journal of phytotherapy and phytopharmacology 17(3-4), 212-218. Rogler, G., 2014. Chronic ulcerative colitis and colorectal cancer. Cancer letters 345(2), 39

235-241. Schreuders, E.H., Ruco, A., Rabeneck, L., Schoen, R.E., Sung, J.J., Young, G.P., Kuipers, E.J., 2015. Colorectal cancer screening: a global overview of existing programmes. Gut 64(10), 1637-1649. Shan, Y.F., Shen, X., Xie, Y.K., Chen, J.C., Shi, H.Q., Yu, Z.P., Song, Q.T., Zhou, M.T., Zhang, Q.Y., 2009. Inhibitory effects of tanshinone II-A on invasion and metastasis of human colon carcinoma cells. Acta pharmacologica Sinica 30(11), 1537-1542. Shapira, M., Ben-Izhak, O., Linn, S., Futerman, B., Minkov, I., Hershko, D.D., 2005. The prognostic impact of the ubiquitin ligase subunits Skp2 and Cks1 in colorectal carcinoma. Cancer 103(7), 1336-1346. Su, C.C., 2012. Tanshinone IIA potentiates the efficacy of 5-FU in Colo205 colon cancer cells in vivo through downregulation of P-gp and LC3-II. Exp Ther Med 3(3), 555-559. Su, C.C., Chen, G.W., Kang, J.C., Chan, M.H., 2008a. Growth inhibition and apoptosis induction by tanshinone IIA in human colon adenocarcinoma cells. Planta medica 74(11), 1357-1362. Su, C.C., Chen, G.W., Lin, J.G., 2008b. Growth inhibition and apoptosis induction by tanshinone I in human colon cancer Colo 205 cells. Int J Mol Med 22(5), 613-618. Sui, J., Li, M., Qian, C., Wang, S., Cheng, Y., Chen, B.P., Wang, D., 2014. Functional analysis of tanshinone IIA that blocks the redox function of human apurinic/apyrimidinic endonuclease 1/redox factor-1. Drug Des Devel Ther 8, 2147-2160. Tian, Y.F., Chen, T.J., Lin, C.Y., Chen, L.T., Lin, L.C., Hsing, C.H., Lee, S.W., Sheu, M.J., Lee, H.H., Shiue, Y.L., Huang, H.Y., Pan, H.Y., Li, C.F., Chen, S.H., 2013. SKP2 overexpression is associated with a poor prognosis of rectal cancer treated with chemoradiotherapy and represents a therapeutic target with high potential. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine 34(2), 1107-1117. Timmerbeul, I., Garrett-Engele, C.M., Kossatz, U., Chen, X., Firpo, E., Grunwald, V., Kamino, K., Wilkens, L., Lehmann, U., Buer, J., Geffers, R., Kubicka, S., Manns, M.P., Porter, P.L., Roberts, J.M., Malek, N.P., 2006. Testing the importance of p27 degradation by the SCFskp2 pathway in murine models of lung and colon cancer. Proceedings of the National Academy of Sciences of the United States of America 103(38), 14009-14014. Tsao, C.K., Cutting, E., Martin, J., Oh, W.K., 2014. The role of cabazitaxel in the treatment of metastatic castration-resistant prostate cancer. Therapeutic advances in urology 6(3), 97-104. 40

Tu, J., Xing, Y., Guo, Y., Tang, F., Guo, L., Xi, T., 2012. TanshinoneIIA ameliorates inflammatory microenvironment of colon cancer cells via repression of microRNA-155. International immunopharmacology 14(4), 353-361. Uddin, S., Ahmed, M., Bavi, P., El-Sayed, R., Al-Sanea, N., AbdulJabbar, A., Ashari, L.H., Alhomoud, S., Al-Dayel, F., Hussain, A.R., Al-Kuraya, K.S., 2008. Bortezomib (Velcade) induces p27Kip1 expression through S-phase kinase protein 2 degradation in colorectal cancer. Cancer research 68(9), 3379-3388. Volpato, M., Perry, S.L., Marston, G., Ingram, N., Cockbain, A.J., Burghel, H., Mann, J., Lowes, D., Wilson, E., Droop, A., Randerson-Moor, J., Coletta, P.L., Hull, M.A., 2016. Changes in plasma chemokine C-C motif ligand 2 levels during treatment with eicosapentaenoic acid predict outcome in patients undergoing surgery for colorectal cancer liver metastasis. Oncotarget 7(19), 28139-28150. Wang, L., Hu, T., Shen, J., Zhang, L., Chan, R.L., Lu, L., Li, M., Cho, C.H., Wu, W.K., 2015. Dihydrotanshinone I induced apoptosis and autophagy through caspase dependent pathway in colon cancer. Phytomedicine : international journal of phytotherapy and phytopharmacology 22(12), 1079-1087. Wang, L., Liu, A., Zhang, F.L., Yeung, J.H., Li, X.Q., Cho, C.H., 2014. Evaluation and SAR analysis of the cytotoxicity of tanshinones in colon cancer cells. Chinese journal of natural medicines 12(3), 167-171. Wang, X., Morris-Natschke, S.L., Lee, K.H., 2007. New developments in the chemistry and biology of the bioactive constituents of Tanshen. Medicinal research reviews 27(1), 133-148. Wei, Z., Jiang, X., Qiao, H., Zhai, B., Zhang, L., Zhang, Q., Wu, Y., Jiang, H., Sun, X., 2013. STAT3 interacts with Skp2/p27/p21 pathway to regulate the motility and invasion of gastric cancer cells. Cellular signalling 25(4), 931-938. Wu, C.Y., Cherng, J.Y., Yang, Y.H., Lin, C.L., Kuan, F.C., Lin, Y.Y., Lin, Y.S., Shu, L.H., Cheng, Y.C., Liu, H.T., Lu, M.C., Lung, J., Chen, P.C., Lin, H.K., Lee, K.D., Tsai, Y.H., 2017a. Danshen improves survival of patients with advanced lung cancer and targeting the relationship between macrophages and lung cancer cells. Oncotarget. Wu, C.Y., Hsieh, C.Y., Huang, K.E., Chang, C., Kang, H.Y., 2012. Cryptotanshinone down-regulates androgen receptor signaling by modulating lysine-specific demethylase 1 function. International journal of cancer. Journal international du cancer 131(6), 1423-1434. Wu, C.Y., Yang, Y.H., Lin, Y.Y., Kuan, F.C., Lin, Y.S., Lin, W.Y., Tsai, M.Y., Yang, J.J., Cheng, Y.C., Shu, L.H., Lu, M.C., Chen, Y.J., Lee, K.D., Kang, H.Y., 2017b. Anti-cancer effect of danshen and dihydroisotanshinone I on prostate cancer: targeting the crosstalk between macrophages and cancer cells via inhibition of the STAT3/CCL2 41

signaling pathway. Oncotarget. Yang, J., Lv, X., Chen, J., Xie, C., Xia, W., Jiang, C., Zeng, T., Ye, Y., Ke, L., Yu, Y., Liang, H., Guan, X.Y., Guo, X., Xiang, Y., 2015. CCL2-CCR2 axis promotes metastasis of nasopharyngeal carcinoma by activating ERK1/2-MMP2/9 pathway. Oncotarget. Yang, Q., Huang, J., Wu, Q., Cai, Y., Zhu, L., Lu, X., Chen, S., Chen, C., Wang, Z., 2014. Acquisition of epithelial-mesenchymal transition is associated with Skp2 expression in paclitaxel-resistant breast cancer cells. British journal of cancer 110(8), 1958-1967. Yang, Y.H., Chen, W.C., Tsan, Y.T., Chen, M.J., Shih, W.T., Tsai, Y.H., Chen, P.C., 2015. Statin use and the risk of cirrhosis development in patients with hepatitis C virus infection. J Hepatol 63(5), 1111-1117. Yashiro, M., 2014. Ulcerative colitis-associated colorectal cancer. World journal of gastroenterology : WJG 20(44), 16389-16397. Yoshimura, T., Liu, M., Chen, X., Li, L., Wang, J.M., 2015. Crosstalk between Tumor Cells and Macrophages in Stroma Renders Tumor Cells as the Primary Source of MCP-1/CCL2 in Lewis Lung Carcinoma. Front Immunol 6, 332.

42

43

44

45

46

47

48

49

Graphical Abstract

50