Development of APTES-Decorated HepG2 Cancer Stem Cell

Nov 16, 2016 - of APTES, which react with the amino groups on cell membranes to form a covalent bond. .... the anti-CSC effect of the screened compone...
3 downloads 0 Views 1MB Size
Subscriber access provided by Fudan University

Article

Development of APTES-decorated HepG2 cancer stem cell membrane chromatography for screening active components from Salvia miltiorrhiza Xuan Ding, Yan Cao, Yongfang Yuan, Zhirong Gong, Yue Liu, Liang Zhao, Lei Lv, Guoqing Zhang, Dongyao Wang, Dan Jia, Zhen Yu Zhu, zhanying hong, Yifeng Chai, and Xiaofei Chen Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b02709 • Publication Date (Web): 16 Nov 2016 Downloaded from http://pubs.acs.org on November 18, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Development

of

APTES-decorated

HepG2

cancer stem cell membrane chromatography for screening

active

components

from

Salvia

miltiorrhiza ‡

§

§

§

Xuan Ding†, Yan Cao†, Yongfang Yuan , Zhirong Gong†, Yue Liu†, Liang Zhao , Lei Lv , Guoqing Zhang , Dongyao Wang†, Dan jia†,Zhenyu Zhu†,Zhanying Hong†, Xiaofei Chen†*, Yifeng Chai†* †

School of Pharmacy, Second Military Medical University, No. 325 Guohe Road, Shanghai 200433, PR

China ‡

Department of Pharmacy, Shanghai 9th People’s Hospital, No. 280 Mohe Road, Shanghai 201999, PR

China §

Department of Pharmacy, Eastern Hepatobiliary Surgery Hospital, No. 225 Changhai Road, Shanghai

200438, PR China

ABSTRACT Cell membrane chromatography (CMC) is an ideal method for screening potential active components acting on target cell membranes from a complex system, such as herbal medicines. But due to the decay and falling-off of membranes, CMC column suffers from short life span and low reproducibility. This has greatly limited the application of this model, especially when the cell materials are hard to obtain. To solve this problem, a novel type of (3-aminopropyl) triethoxysilane (APTES)-decorated silica gel was employed. The silica gel was decorated with aldehydes with the help of APTES, which react with the amino groups on cell membranes to form covalent bond. In this way, cell membranes were immobilized on the surface of silica gel, so it is not easy for membranes to fall off. According to our investigation, column life of APTES-decorated group was prolonged to more than 12 days, while control group showed a sharp decline in column efficiency in the first 3 days. To verify this model, a novel APTES-decorated HepG2 cancer stem cell membrane chromatography (CSCMC) was established and applied in a comprehensive two-dimensional chromatographic system to screen potential active components in Salvia miltiorrhiza. As a result, tanshinone IIA, cryptotanshinone and dihydrotanshinone I were retained on this model and proved to be effective on HepG2 cancer stem cell by the following cell proliferation and apoptosis assay, with IC50 of 10.30 µM, 17.85 µM and 2.53 µM respectively. This improvement of CMC can significantly prolong its column life span and broad the range of its application, which is very suitable for making invaluable or hard-to-obtain cell materials, such as stem cells, into CMC for specific drug screening.

1

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 16

INTRODUCTION Cell membrane chromatography (CMC) combines the advantages of biomaterials and chromatography, making it possible to screen active compounds from a complex system online1-3. Its theory is that the cell membrane is immobilized on the stationary phase and the sample solution of a complex system is injected into the CMC system to investigate the retention behavior of all the compounds by a mass-spectrometry or other kinds of detectors. The stronger the compound is retained, the more likely it is to be active. This technology is similar with other affinity chromatography as reported4-7. Our group has successfully developed a series of CMC models, including a comprehensive two-dimensional CMC/C18 system1,8. Traditionally, the binding force between cell membranes and silica gel in CMC model is hydrophobic interaction3,9. This interaction is relatively weak and unstable in comparison to covalent binding, which makes it easy for membranes to fall off from the silica gel. As a result, CMC suffers from short column life span and low column efficiency1,8,10. Those problems have greatly limited the application of this practical model. Other affinity chromatography also suffers from similar problems at first, but these problems were finally solved by immobilizing proteins onto stationary phase with modification of stationary phase to form covalent bond with proteins11. On account of this, modification of silica gel was considered. As we know, cell membranes are rich in phospholipids12,13. Some kinds of phospholipids such as phosphatidyl-ethanolamine and phosphatidyl-serine have a free amino group at one end. There have been reports to immobilize nanoparticles, which is rich in phospholipids as well, in the capillary under the help of (3-aminopropyl)triethoxysilane (APTES)14-16. In view of this, we considered to decorate the silica gel with aldehydes under the help of APTES as reported to form a free aldehyde group at one end14, which react with the amino group on cell membranes to form a covalent bond. In this way, the binding force between cell membranes and silica gel is replaced as covalent bond rather than hydrophobic interaction. As a result, it becomes not easy for membranes to fall off. This novel CMC model is very suitable for biological material which is of high value or is hard to obtain, such as cancer stem cells (CSCs). Cancer stem cells are a bunch of special tumor cells which are capable of self-renewal and play an important role in the recurrence, metastasis and migration of cancer17. Killing CSCs will significantly improve the cure rate of cancer and simultaneously reduce the recurrence rate18. But it’s generally tedious and time-consuming to separate and culture CSCs. Therefore, our newly established APTES-decorated CMC model is very suitable for CSCs. It is reported that more than 70% of all drugs approved from 1981 and 2006 were either derived from or structurally similar to nature based compounds19. Thus, Salvia miltiorrhiza, a famous herbal medicine with definite clinical effects, was chosen as an ideal complex system. Additionally, a comprehensive two-dimensional chromatographic system developed by our group was also applied to improve the resolution of CMC8. Based on these conditions, a covalently APTES-decorated HepG2 cancer stem cell membrane chromatography/RP-18e monolithic column/time-of-flight mass spectrometry system was established and the retained components in this system were definitely identified and confirmed. At last, pharmacological verification trials, such as cell proliferation, apoptosis experiments and molecular docking 2

ACS Paragon Plus Environment

Page 3 of 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

assays, were conducted to verify the anti-CSC effect of the screened components. This system will be an effective tool to screen target components from complex systems acting on cell membrane receptors from invaluable biological materials.

EXPERIMENTAL SECTION Reagents and instruments. Silica gel (5 µm, 200 Å) was obtained from Qingdao Meigao Chemical Co., Ltd. (Qingdao, China) and was activated at 120 °C before using. Gefitinib (GFT), Dexamethasone (DXM), salvianolic acid B, tanshinone IIA, cryptotanshinone and dihydrotanshinone I (purity>99.5%) were purchased from Meilun Pharmaceutical Co., Ltd. (Dalian, China). Salvia miltiorrhiza was purchased from Leiyunshang Pharmacy Co., Ltd. (Shanghai, China). (3-aminopropyl) triethoxysilane (APTES, purity>99.5%) and glutaric dialdehyde (GDD, purity>99.5%) were purchased from Sigma Chemical Co. (Missouri, USA). Dulbecco’s modified Eagle’s medium (DMEM),Dulbecco's Modified Eagle Media: Nutrient Mixture F-12 (DMEM-F12) and Phosphate buffer saline (PBS) were purchased from Hyclone (Thermo Fisher). Fetal bovine serum (FBS), B27, bFGF, EGF, and insulin were obtained from Gbico Life Technology Co. (Australia). Dimethyl sulfoxide (DMSO), penicillin streptomycin and trypsin were purchased from Gibco-BRL Co. (Rockville, MD, USA). Ultrapure water was prepared by a Milli-Q Academic A10 water purification system (Millipore, Bedford, MA, USA). Acetonitrile (Merck, Germany) and formic acid (Merck, Germany) were of HPLC grade. Other reagents were of analytical grade. The electronic balance used was AND HA-202M (Japan). The centrifuge was HITACHI CR21GIII from Hitachi Co., Ltd. (Japan). CCK-8 kit and Cell apoptosis kit were purchased from Beyotime Co. (Shanghai, China). The HPLC system was Angilent-1100 from Agilent Technologies Co., Ltd. (California, America) coupled with an Angilent-6200 mass spectrometer and a Mass Hunter LCMS workstation. The ultrasonic processor was JY92-IIN from Scientz Biotechnology (Ningbo, China). An electronically controlled 10-port dual-position valve (MXP9960-000, Rheodyne, Rohnert park, CA, USA) equipped with a CMC column and a Chromolith Performance RP-18e monolithic silica column (100 mm × 4.6 mm I.D., Merck, Darmstadt, Germany) was used. Preparation of samples and standard solutions. Salvia miltiorrhiza was harvested from Sichuan, China in February, 2015 by the Leiyunshang Pharmacy Co., Ltd. and identified by prof. Baokang Huang from School of Pharmacy, Second Military Medical University. It was firstly disposed into powder. Then the powder was refluxed in 50 % ethanol at 80 °C for 1 h and then condensed to 10 mg/mL. The sample was filtered by 0.2 µm filter membrane before analysis. Standard solutions of GFT, DXM, salvianolic acid B, tanshinone IIA, cryptotanshinone and dihydrotanshinone I were prepared by dissolving in DMSO at the concentration of 20 mM respectively before use. Cell culture. HepG2 was selected as a representative cancer cell series in this paper, which was purchased from Cell Bank of Shanghai Branch of Chinese Academy of Sciences. The cells were maintained in Dulbecco’s minimum essential medium (DMEM) supplemented with 10% fetal bovine serum (FBS), in a humidified atmosphere of 5% CO2 at 37 °C. 3

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 16

Tumor spheres were reported to be representative of cancer stem cells20-22. To obtain the tumor spheres of HepG2 cells, HepG2 cells were plated at 1×104 cells/mL in serum-free DMEM-F12, supplemented with 2% (v/v) B27, 10 ng/mL bFGF, 20 ng/mL EGF, 5 mg/mL insulin. Tumor spheres were enzymatically dissociated every six days by incubation in pancreatin for 1 min at 37°C and passaged at 1×104 cells/mL. Cells used in this study were within 5 generations and in the logarithmic phase of growth. Animals. All nude mice (female, 18-20 g, 6-8 weeks) were purchased from Shanghai Experimental Animal Center of Chinese Academic of Sciences (Shanghai, China). Mice were placed in a pathogen-free environment and allowed to acclimate for a week before use. All procedures were performed in accordance with guidelines of the Committee on Animal of the Second Military Medical University (Shanghai, China). Identification of CSCs in HepG2 tumor spheres. Liver CSCs could be enriched by a variety of phenotypes such as CD13323, however, these phenotypes showed relatively lower specificity and sensitivity in identifying liver CSCs24. In fact, there has been no generally accepted marker for liver CSCs. Therefore an approach called tumor sphere formation is considered to be practical to isolate CSCs25. In this study, cultured HepG2 tumor spheres were demonstrated to possess the characteristics of liver CSCs, by using the well-defined methods, including in tumor sphere formation ability, tumor formation ability and the surface markers (CD133). More importantly, the tumor spheres could be cultured and propagated. For tumor sphere formation ability assay, a certain quant of HepG2 cells or single cells dissociated from the third generation of HepG2 tumor spheres were planted on a six-well plate. After 6 days of incubation in serum free DMEM, the number of tumor spheres were counted and compared between each group (n = 3). For tumor formation ability assay, a series of quant of HepG2 cells or single cells dissociated from the third generation of HepG2 tumor spheres (mixed with 100 mL of Matrigel) were implemented into the mammary fat pad of mice to investigate the tumor formation of each group (n = 5). The surface markers (CD133) of the tumor spheres were analyzed by flow cytometry (FCM). Briefly, the third generation of tumor spheres were enzymatically dissociated into single cells, washed with PBS twice and then resuspended. The cells were simultaneously stained with 20 µL PE labeled mouse anti-human CD133 antibody at 4°C for 30 min in the staining buffer. Then the cells were washed with cold washing buffer for three times. Finally, the cells were resuspended in 500 µL of cold PBS, and assayed by a FACS Calibur CM (Becton Dickinson, USA) equipped with BD Cell Quest software. Synthesis of APTES-decorated stationary phase. The reaction scheme is shown in Fig. 1. Firstly, the silica gel was decorated with APTES to obtain an amino group on the surface. In brief, 0.2 g degassed silica gel was mixed with 0.1 mL APTES in 10 mL toluene under inert argon atmosphere at 110 °C for 12h. Then the sample was dried and suspended in 100 mL glutaraldehyde(5 wt%) diluted in methanol and shaken for 2h under room temperature for the binding of glutaraldehyde onto APTES14,26. At last, the other end of glutaraldehyde would be able to link to cell membranes by reacting with the abundant amino groups on the membranes 4

ACS Paragon Plus Environment

Page 5 of 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

after 5 min vortex in vacuum and 12 h incubation at 4°C.

Fig. 1. Synthesis of APTES-decorated silica gel and its reaction with cell membranes. Preparation of HepG2/CSCMC columns. HepG2 cancer stem cell membrane chromatographic (HepG2/CSCMC) columns were prepared according to previously reported methods2 using the APTES-decorated silica gels as mentioned above. Briefly, 3.5×107 cells were harvested and then washed three times with PBS by centrifuging at 110×g for 10 min. PBS was then added to produce a cell suspension. The cells were disrupted by an ultrasonic processor. The resulting homogenate was centrifuged at 1,000 × g for 10 min. The pellet was discarded and the supernatant was centrifuged at 12,000 × g for 20 min. The precipitation was then suspended in 5 mL PBS. Cell membrane stationary phase (CMSP) was prepared by the adsorption of cell membrane suspension on 0.04 g APTES-decorated silica gel under vacuum and agitation conditions. All the procedures above were implemented at 4 °C. After overnight-incubation for 12 h, the CMSP was washed three times with PBS by centrifuging at 110 × g for 5 min. The pellet was suspended in PBS and packed into the column (10 mm × 2 mm I.D., purchased from Dalian Replete Science and Technology Co., Ltd.) by an LC pump (Waters 996) with PBS. The column is equilibrated 1 h at the flow rate of 0.2.mL min−1 10 5

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 16

mM Ammonium acetate and the temperature of 37 °C until stable column pressure and baseline have been obtained. The CMC columns were stored in ammonium acetate at 4 °C. APTES-decorated 2D HepG2/CSCMC analysis. As our group reported before1,8, the APTES-decorated 2D HepG2/CSCMC system was performed on an Agilent 1200 series HPLC system consisting of a unitary and a binary solvent delivery systems, a thermostatically controlled column apartment, an on-line degasser and an autosampler, controlled by Agilent MassHunter Workstation (Agilent Technologies, Palo Alto, CA, USA). A HepG2/CSCMC column (10 × 2 mm i.d., 5 µm) was applied as the first dimensional column and the mobile phase was 10 mM ammonia acetate delivered at 0.2 mL.min−1. For the second dimension separation, a Chromolith Performance RP-18e monolithic silica column (100 mm × 4.6 mm I.D., Merck, Darmstadt, Germany) was used and the mobile phase, composed of solvent A (0.1% formic acid) and solvent B (acetonitrile) and delivered at 3.5 mL.min−1 by a linear gradient elution program. The program was set as follows: 0-4 min, from 10% B to 30% B; 4-9 min, from 30% B to 75% B; 9-9.01 min, from 75% B to 10% B; 9.01-10 min, 10% B. Detailed steps of operation of the 2D system were stated in our group’s previous researches1,8. Cell proliferation and cell apoptosis assay. Cell proliferation assay kits for CCK-8 (Beyotime Co., Shanghai, China) were used according to the manufacturer’s instructions. GFT was used as a positive control according to our previous researches2,8. After 24 h incubation, cells were treated with GFT, salvianolic acid B, tanshinone IIA, cryptotanshinone and dihydrotanshinone I at various concentrations for 48 h. Then, 10 µL of CCK-8 solution was added to each well and incubated at 37 °C for 1.5 h. Bubbles were removed with a hair dryer prior to reading the absorbance at 450 nm on a microplate absorbance reader (Bio-RAD instruments, USA). Cell apoptosis was assayed by flow cytometry with the Annexin V-FITC Apoptosis Detection kit (BD Pharmingen™ CA, USA). 5 × 105 cells were plated on each well of six-well plate and treated with GFT, salvianolic acid B, tanshinone IIA, cryptotanshinone and dihydrotanshinone I at the dose of 20 µM for 48 h. Then, cells were harvested, washed twice in PBS. Annexin V/FITC was then added. After incubation for 10 min at room temperature in the dark, the cells were washed and resuspended; propidium iodide was then added to a final concentration of 1 mg/l. Stained cells were analyzed using a FACS Calibur instrument (Becton Dickinson, Mountain View, CA, USA). Molecular Docking. The crystal structure of VEGFR2 was downloaded from the Protein Data Bank with PDB code 3VHE27. Crystal waters and co-crystalized ligands were removed, and hydrogen atoms were added to the protein according to the protonation states of the chemical groups at pH 7.0. The binding site was determined to include any protein atom within 4 Å of any atom of the complex ligand. Dihydrotanshinone, tanshinone IIA, and cryptotanshinone were docked into the defined catalytic site by the program LeDock (http://lephar.com), which performs an exhaustive search of position, orientation and conformation of the ligand28.

6

ACS Paragon Plus Environment

Page 7 of 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

RESULTS AND DISCUSSION Identification of tumor spheres as CSCs. As shown in Fig. 2C, cells isolated from the third generation of HepG2 tumor spheres exhibited higher ratio of tumor formation ability compared with those isolated from HepG2 cells, indicating that HepG2 tumor spheres possess more ability to form tumor spheres again than HepG2 cells. As shown in Fig. 2D, the third generation of HepG2 tumor spheres induced tumors with a 100% incidence when the cell account was larger than or equal to 2×103. On the contrary, HepG2 cells induced tumors with a 100% incidence only when the cell account was larger than or equal to 2×105,100 times more than that of HepG2 tumor spheres. In summary, HepG2 tumor spheres cells possessed much enhanced tumor formation ability than HepG2 cells. According to the FCM (Fig. 2A and 2B), 24.09% of the cells in HepG2 tumor spheres showed a CD133 phenotype, which is considered to be the most common type of HepG2-CSCs, compared with 4.67% from HepG2 cells. Therefore, the third generation of tumor spheres of HepG2 was selected as representative of HepG2-CSCs, identified by tumor sphere formation ability, tumor formation ability and the surface markers (CD133).

Fig. 2. Identification of tumor spheres as cancer stem cells: (A) typical flow cytometry results of CD133 marker for HepG2 tumor spheres, (B) typical flow cytometry results of CD133 marker for HepG2 cells, (C) tumor sphere formation assay results (n = 3) and (D) tumor formation ability results (n = 5).

7

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 16

Fig. 3. Investigation of CMC columns: (A) Comparison of results of column efficiency from HepG2 control columns and HepG2-APTES-modified columns, (B) retention time of GFT on control columns and APTES-decorated columns prepared with adequate cells within 12 days (n = 6), (C) selectivity evaluation of the APTES-decorated 2D HepG2/CSCMC system (1 stands for DXM as a negative control ligand and 2 stands for GFT as a positive ligand in A and C). Reproducibility and life span of APTES-decorated CMC columns. The cell membrane chromatographic columns gradually lose activity during use, resulting in low reproducibility and limited life span within 3 days. It has greatly prevented the CMC from being widely used. To solve this problem, we have previously used a post-preparation treatment by a chemical compound called paraformaldehyde, which successfully strengthened the column reproducibility and prolonged the life span to about 6 days2. But this procedure added the difficulty of column preparation and might have influence on the bioactivity of cell membranes to some extent. So we have diverted our attention to the decoration of the stationary phase, expecting to find a solution by the modification of gels. As known, cell membranes are made of phospholipids, which provide a great deal of amino groups. Bearing this in mind, we have synthetized a kind of silica gel decorated with APTES. The APTES-decorated gels are rich in aldehyde groups, which can be covalently bound to the amino groups from the cell membrane. Since HepG2-CSCs are hard to obtain and investigation of APTES-decorated silica gel 8

ACS Paragon Plus Environment

Page 9 of 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

requires large amounts of cell membranes, we used HepG2 cells instead of HepG2-CSCs as an evaluation model to investigate the effect of the decorated silica gel. As shown in Fig. 3A, to verify the improved column efficiency for APTES-decorated silica gel, 1.0 ×107 cells were used to prepare HepG2 cell membrane columns made from the APTES-decorated and non-decorated silica gel (n = 6). Results showed that retention time of GFT on APTES-decorated CMC columns is about 21 min, while the retention time on control group columns is only about 10 min. This indicates that APTES-decorated group showed higher column efficiency than the non-decorated group. To further validate the column life span, two groups of HepG2/CMC columns were made from APTES-decorated silica gel and non-decorated silica gel respectively under optimized condition. Retention time of marker ligand, GFT as reported before2,8,29, was investigated for 12 days with 6 injections for each day (n = 6). As shown in Fig 3B, this treatment procedure has greatly strengthened the immobilization of cell membranes on the APTES-decorated silica gel by covalent bonding rather than traditionally used procedures by hydrophobic interaction, prolonging column life span to at least 12 days and enhancing the reproducibility to within 10 % compared with the previously reported result of 20 % for the first 3 day use1,8,10, during when the column efficiency falls most. Application of the 2D HepG2/CSCMC system. Selectivity of the 2D HepG2/CSCMC system was evaluated first. GFT (epidermal growth factor receptor antagonist) and DXM (hormones), selected as the positive and negative control drugs respectively as our group reported before2,8,29, were used to confirm the selectivity of the APTES-decorated 2D HepG2/CSCMC system. GFT showed strong retention behavior, reaching peak at about 17 min. While DXM barely retained on this CMC model, reaching peak at about 2 min. This indicates satisfactory selectivity of this system as shown in Fig. 3C. The 2D HepG2/CSCMC system was then applied into the screening of potential active compounds from Salvia miltiorrhiza. As shown in Fig. 4A, totally 12 components were directly observed in the 2D contour spectrum and were credibly identified with accurate mass data and isotope abundant fragment information provided by TOFMS1,8,29,30(shown in Table 1). Tanshinone IIA, methyl tanshinonate, tanshinone V, cryptotanshinone, dihydrotanshinone I, 3α-hydroxytanshinone IIA and isotanshinone IIB were significantly retained on HepG2/CSCMC column, while salvianolic acid B was barely retained. As we know, herbal medicine usually contains hundreds of components. In consideration of screening efficiency and speed, a total of 10 min running time was conducted for the second dimensional monolithic chromatography, resulting in overlaps among some different components. Most of them were barely retained in the first CMC dimension, which were considered as non-active components in this experiment. A mixed standard solution containing four typical components (salvianolic acid B for negative and tanshinone IIA, cryptotanshinone and dihydrotanshinone I for positive at 10 mM each) was used to verify the screening results. As shown in Fig. 4B, those four components exhibited expected retention characteristics. Thus the major potential active components were finally identified, which were in accordance with the tentatively identification results of TOFMS.

9

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 16

Fig. 4 Typical 2D chromatography plots of (A) the sample solution retaining results of 2D HepG2/CSCMC system and (B) the standard solution retaining results (information of components 1 to 12 were listed in Table 1). Table 1. Retention components of Salvia miltiorrhiza on 2D HepG2/CMC system identified by TOFMS Peak

Identification

tR(CMC, min)

tR(RP-18e, min)

number 1

2

a

Tanshinone IIA

Tanshinone I

1-30

b

1-5

8.0

7.1

m/z

Abundance

Expected

Detected

Error(ppm)

match(%)

295.1329

295.1329

0

97.02

C19H18O3

([M+H]+)

([M+H]+)

277.0859

277.0857

-0.72

98.56

C18H12O3

1.34

97.92

C19H20O3

0.59

97.86

C20H18O5

2.54

95.08

C19H22O4

-0.72

99.61

C18H14O3

0.30

98.25

C19H18O5

-1.93

93.14

C19H18O4

2.99

97.72

C36H30O16

-3.08

98.00

C10H10O4

0.28

84.02

C18H14O8

-7.19

92.16

C7H6O3

+

3

Cryptotanshinonea

1-23

7.0

+

([M+H] )

([M+H] )

297.1485

297.1489

+

4

5

Methyl tanshinonate

Tanshinone V

1-25

2-30 b

6.7

6.5

+

([M+H] )

([M+H] )

337.1227

337.1229

([M+H]+)

([M+H]+)

315.1583

315.1591

+

6

Dihydrotanshinone Ia

13-30b

6.1

+

([M+H] )

([M+H] )

279.1016

279.1014

+

7

8

3α-hydroxytanshinone IIA

Isotanshinone IIB

3-29

1-28

5.7

5.1

+

([M+H] )

([M+H] )

327.1227

327.1228

([M+H]+)

([M+H]+)

311.1278

311.1272

+

+

([M+H] ) 9

Salvianolic acid Ba

1-3

3.1

([M+H] )

736.1872

736.1894 +

10

11

Caffeic acid methyl ester

Prolithospermic acid

1-7

1-2

2.2

2.0

Protocatechualdehyde

1-2

1.1

+

([M+NH4] )

([M+NH4] )

195.1088

195.1082

([M+H]+)

([M+H]+)

359.0761

359.0762

+

12

Formula

+

([M+H] )

([M+H] )

139.039

139.038 +

([M+ H] )

+

([M+ H] )

a. Confirmed by authentic standard compounds. 10

ACS Paragon Plus Environment

Page 11 of 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

b. Peak that was not completely flushed out by 1st-CMC column within 30 min.

Cell proliferation and cell apoptosis assay. Previous studies have shown that tanshinone IIA, cryptotanshinone and dihydrotanshinone I could induce cell proliferation of the HepG2 cells31,32, but very limited data were available of their effect on HepG2-CSCs. Results of cell proliferation assay were in accordance with our screening data, as shown in Fig. 5A and 5B. Dihydrotanshinone I exhibited the strongest inhibition effect on HepG2-CSCs in a dose-dependent manner, followed by tanshinone IIA and cryptotanshinone. But salvianolic acid B showed little inhibition effect. In addition, a traditionally used EGFR-targeted drug GFT was selected as a positive control. IC50 of GFT in HepG2-CSCs was 45.74 µM higher than that in HepG2 cells (24.23 µM), indicating that GFT showed lower bioactivity in cancer stem cells. This is in accordance with our CMC screening results that peak reaching time of GFT on APTES-decorated HepG2/CMC columns was longer than that on APTES-decorated HepG2/CSCMC columns. On contrary, as shown in Fig. 5C, IC50 of tanshinone IIA, cryptotanshinone and dihydrotanshinone I was lower in HepG2-CSCs than that in HepG2 cells, indicating they have a better bioactivity in cancer stem cells. Thus, tanshinone IIA, cryptotanshinone and dihydrotanshinone I were identified as potentially anti-CSC components and showed better bioactivity in HepG2-CSCs than that in HepG2 cells.

Fig. 5. Cell proliferation results of (A) HepG2-CSCs and (B) HepG2 cells, statistical differences were estimated with Student’s t-test(*p < 0.05 was taken as statistically significant and **p < 0.01 was considered as dramatically significant vs. the negative control, n = 5). (C) Comparison of IC50 of tanshinone IIA, cryptotanshinone, dihydrotanshinone I and GFT between HepG2 (53.050 µM, 34.140 µM, 3.661 µM and 24.230 µM respectively) and HepG2-CSCs(10.300 µM, 17.850 µM, 2.534 µM and 45.740 µM respectively). To further identify the effect of screened drugs and to assess whether the growth inhibition in HepG2-CSCs was related to apoptosis, cell apoptosis assay was carried out. As shown in Fig. 6A and 6B, after exposure to salvianolic acid B (negative control), tanshinone IIA, cryptotanshinone and dihydrotanshinone I and GFT (positive control) at the dose of 20 µM for 48 h, the percentage of apoptotic cells of drug-treated groups significantly increased 11

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 16

compared with control group except for the salvianolic acid B group, demonstrating that tanshinone IIA, cryptotanshinone and dihydrotanshinone I, which were screened from 2D HepG2/CSCMC system, can effectively induce apoptosis in HepG2-CSCs in spite of their different differentiation states. But percentage of apoptotic cells by GFT has dropped from 50.32% (HepG2) to 11.93% (HepG2-CSCs) as shown in typical flow cytometry results in Fig. 6A. In addition, In addition, Fig. 6B indicates that tanshinone IIA, cryptotanshinone and dihydrotanshinone I showed better bioactivity in HepG2-CSCs compared with HepG2 cells.

Fig.6 (A) Typical flow cytometry results of salvianolic acid B (negative control), tanshinone IIA, cryptotanshinone, dihydrotanshinone I and GFT (positive control) on HepG2-CSCs (left) and HepG2 cell (right) (B) Cell apoptosis and death rate of HepG2 cells and HepG2-CSCs (n = 3), with **p