Multipoint Costriking Nanodevice Eliminates Primary Tumor Cells and Associated-Circulating Tumor Cells for Enhancing Metastasis Inhibition and Therapeutic Effect on HCC

Eliminating primary tumor ("roots") and inhibiting associated-circulating tumor cells (associated-CTCs, "seeds") are vital issues that need to be urgently addressed in cancer therapy. Associated-CTCs, which include single CTCs, CTC clusters, and CTC-neutrophil clusters, are essential executors in metastasis and the cause of metastasis-related death in cancer patients. Herein, a "roots and seeds" multipoint costriking nanodevice (GV-Lipo/sorafenib (SF)/digitoxin (DT)) is developed to eliminate primary tumors and inhibit the spread of associated-CTCs for enhancing metastasis inhibition and the therapeutic effect on hepatocellular carcinoma (HCC). GV-Lipo/SF/DT eliminates primary tumor cells by the action of SF, thus reducing CTC production at the roots and improving the therapeutic effect on HCC. GV-Lipo/SF/DT inhibits associated-CTCs effectively via the enhanced identification and capture effects of glypican-3 and/or vascular cell adhesion molecule 1 (VCAM1) targeting, dissociating CTC clusters using DT, blocking the formation of CTC-neutrophil clusters using anti-VCAM1 monoclonal antibody, and killing CTCs with SF. It is successfully verified that GV-Lipo/SF/DT increases the CTC elimination efficiency in vivo, thus effectively preventing metastasis, and shows enhanced antitumor efficacy in both an H22-bearing tumor model and orthotopic HCC models. Overall, the "roots and seeds" multipoint costriking strategy may open a new cancer treatment model for the clinic.

dissolved in PBS (pH 7.4) and stirred for 8 h at room temperature. The product was purified by dialysis (MWCO 5000 Da, Millipore) for 2 days. After lyophilization, the product was verified by 1 H NMR (Avance™ DPX-300; Bruker BioSpin GmbH, Rheinstetten). The DSPE-PEG 2000 -anti-VCAM1 mAb was synthesized by the same method except that the anti-GPC3 mAb was replaced with the anti-VCAM1 mAb.
Expression of GPC3 and VCAM1. The expression of GPC3 and VCAM1 in the Hepa1-6 cell line and H22 cell line was analyzed. Briefly, Hepa1-6 cells were seeded in a confocal dish (1×10 5 cells/well) and incubated overnight. Then, fresh PBS (pH 7.4) containing a FITC-labelled anti-GPC3 mAb (200 ng/mL) was added and incubated for 0.5 h on ice. After the incubation, the cells were fixed in 4% paraformaldehyde and then stained with 4′,6-diamidino-2-phenylindole (DAPI) for 10 min at room temperature. Finally, the dish was examined under CLSM (LSM 780,Carl Zeiss). To quantify the expression of GPC3, Hepa1-6 cells were digested after incubation with the FITC-labelled anti-GPC3 mAb, centrifuged, resuspended in cold PBS (pH 7.4) and detected by a flow cytometer (BD FACSCelesta). To verify the expression of GPC3 in H22 cells, H22 cells were added to a 2-mL centrifuge tube at a density of 1×10 5 and analyzed using the method described above. The expression of VCAM1 in the Hepa1-6 cell line and H22 cell line was analyzed using the same method, ON) five times. The particle size, polydispersity index (PDI), and zeta potential of GV-Lipo/SF/DT were determined with a Zetasizer Nano ZS90 (Malvern, Worcestershire).
The morphology of GV-Lipo/SF/DT was evaluated by transmission electron microscopy (TEM). The EE% and drug loading (DL%) of SF and DT were calculated using the following equations: DL% = (W loaded drug /W liposomes ) × 100% (1) EE% = (W loaded drug /W total drug ) × 100% ( 2) where W loaded drug represents the amounts of loaded drug in GV-Lipo/SF/DT and W total drug and W liposomes represent the amounts of added drug and liposomes, respectively. All samples were evaluated in triplicate.
To verify that the targeted antibodies were modified on the surface of the liposomes, the anti-GPC3 mAb was labeled with FITC, and the anti-VCAM1 mAb was labeled with RhB.
Then, fluorescent dye-labeled GV-Lipo/SF/DT was imaged by CLSM. The amount of modification by the targeting antibodies was analyzed with BCA Protein Assay Reagent. The modification rates of the anti-GPC3 mAb or the anti-VCAM1 mAb on liposomes were calculated using the following equations: Modification rates (%)=(W modified mAb /W total mAb ) × 100% where W modified mAb represents the amounts of anti-GPC3 mAb or anti-VCAM1 mAb modified on the surface of GV-Lipo/SF/DT and W total drug represent the amounts of added anti-GPC3 mAb or anti-VCAM1 mAb when GV-Lipo/SF/DT prepared. In vitro cytotoxicity assay. Hepa1-6 cells (5000 cells/well) were exposed to blank liposomes (blank Lipo), free SF, free DT, Lipo/SF, Lipo/DT, Lipo/SF/DT or GV-Lipo/SF/DT for 48 h.

Stability of GV-Lipo/SF/DT in different media.
After treatment, 20 μL 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, 5 mg/mL) was added and incubated at 37 °C for 4 h. Then, the medium was removed, and 150 μL DMSO was added. The untreated cells group was set as the control group and PBS group was set as blank group. The cell viability in each group was measured using a microplate reader (Model 680; Bio-Rad, CA). All experiments were carried out in triplicate.
Relative cell viability (%) was calculated using the following equation: where A sample represents the absorbance value of different formulation, A control represents the absorbance value of control group, and A blank represents the absorbance value of blank group.
In vitro targeting assay evaluating GV-Lipo/SF/DT. The targeting ability of the formulation was evaluated in an in vitro cellular uptake experiment. C6 was used as the fluorescence indicator in the cellular uptake experiment. Hepa1-6 cells were seeded overnight in a 12-well plate (1×10 5 cells/well). After discarding the medium, fresh DMEM containing free C6, C6-loaded liposomes (C6-Lipo), anti-GPC3 mAb-modified C6-loaded liposomes (G-Lipo/C6), anti-VCAM1 mAb-modified C6-loaded liposomes (V-Lipo/C6), or anti-GPC3 mAb and anti-VCAM1 mAb dual-modified C6-loaded liposomes (GV-Lipo/C6) (C6 concentration: 200 ng/mL) was added and further incubated for 0.5 h. Then, the cells were washed twice with PBS (pH 7.4) and stained with DAPI for 10 min. Finally, the cells were imaged using an imaging reader (BioTek Cytation 1). To quantify the targeting abilities of the formulations, after treatment with free C6, C6-Lipo, G-Lipo/C6, V-Lipo/C6 or GV-Lipo/C6, the Hepa1-6 cells were digested, centrifuged, re-suspended and detected by FCM. Furthermore, a competitive inhibition experiment was used to evaluate the targeting properties of GV-Lipo/C6. After pre-incubation with excess free anti-GPC3 mAb, anti-VCAM1 mAb, or anti-GPC3 mAb + anti-VCAM1 mAb, GPC3 and/or VCAM1 were saturated. Following the addition of G-Lipo/C6, V-Lipo/C6 or GV-Lipo/C6 into the wells and incubation for 0.5 h, the cells were imaged using an imaging reader and FCM (BD FACSCelesta). Then, the suspended cells were incubated under the action of the rotary viscometer (shear rate: 188 s −1 ) for 0.5 h at 37 °C. After the incubation, the suspended cells were centrifuged to remove the medium and stained with DAPI for 10 min. Finally, the cells were imaged using an imaging reader and quantified by FCM.

In vitro CTC cluster dissociation assay.
To investigate the CTC cluster dissociation ability of DT, free DT and Lipo/DT were resuspended in Hepa1-6 CTC medium (5000 cells/well) at different concentrations (0, 0.5, 1, 1.5, 2, 3, 4, and 5 μM) and incubated for 48 h in 96-well plates. After the incubation, cell morphology was examined with an imaging reader (BioTek Cytation 1). Cytotoxicity was further studied; 20 μL of 5 mg/mL MTT was added and incubated at 37 °C for 4 h. Then, DMSO was added, and cell viability was measured and calculated as described in the "In vitro cytotoxicity assay" subsection.

Intracellular Ca 2+ concentration determination after treatment with DT. To study whether
DT influences the intracellular Ca 2+ concentration, the intracellular Ca 2+ concentration in Hepa1-6 cells was measured by BBcellProbe ® F03 (Shanghai BestBio Biotechnology Co., Ltd.). Briefly, Hepa1-6 cells were incubated overnight in a 12-well plate at a density of 1×10  Isolation of neutrophils from mouse peripheral blood. Neutrophils were obtained from the whole blood of C57BL/6 mice with a Percoll gradient centrifugation method using a peripheral blood neutrophil isolation kit (Solarbio, PRC). Blood samples were collected in heparin tubes on a sterile test bench and diluted with PBS. Reagent A (4 mL), reagent C (2 mL) and whole blood diluent (2 mL) were added to 15-mL centrifuge tubes and centrifuged at 800 g for 30 min at room temperature. After centrifugation, the neutrophil layer was carefully collected and washed twice with PBS (250 g, 10 min) to obtain high-purity neutrophils.
Lipopolysaccharide (LPS, 100 μg/mL) was used to activate neutrophils in serum-free 1640 medium for 6 h.

Prevention of CTC-neutrophil cluster formation in vitro. Neutrophils co-cultured with GFP
Hepa1-6-luc cells were used to evaluate the ability of the anti-VCAM1 mAb to prevent the formation of CTC-neutrophil clusters. GFP Hepa1-6-luc cells (4×10 5 ) and neutrophils ( 8×10 5 ) were seeded in a confocal dish and co-incubated overnight. The following groups were included: the GFP Hepa1-6-luc cell + neutrophil co-incubation group, anti-VCAM1 mAb + GFP Hepa1-6-luc cell + neutrophil co-incubation group and V-lipo + GFP Hepa1-6-luc cell +   and day 20. Two-month survival curves were recorded after treatment.

In vivo pulmonary metastasis inhibition study.
The inhibitory effect of GV-Lipo/SF/DT on pulmonary metastasis in vivo was evaluated in KM mice i.v. injected with H22 cells. H22 cells (8 × 10 6 ) were i.v. injected into mice, which were then treated with different formulations as described in the "Antitumor efficacy of GV-Lipo/SF/DT in H22-bearing tumor model" subsection. The lungs were collected, imaged, and stained with H&E on the 21st day. An immunofluorescence assay was also conducted to characterize whether anti-VCAM1 mAb would effect the number of neutrophils in pulmonary metastatic nodules.
The sections of lung were stained as described in "Prevention of CTC-neutrophil cluster formation in vitro". The number of pulmonary metastasis nodules was counted. Inhibition rate (%) on pulmonary metastasis was calculated using the following equation: where N sample represents the number of pulmonary metastasis nodules after treated with different samples, N NS represents the number of pulmonary metastasis nodules of NS group.
Hemolysis assay. Samples were mixed with a red blood cell suspension (2%) and incubated at 37 °C. After centrifugation, the absorbance of hemoglobin was determined by a UV-vis spectrophotometer (576 nm). The hemolysis rates of different samples were calculated.
A x , A 1 , and A 0 represent the absorbance of the samples, positive control, and negative control, respectively.

Statistical analysis.
Statistical analyses were calculated with GraphPad Prism 6. Student's t-tests were used for comparisons between two groups according to the data distribution.
One-way analysis of variance (ANOVA) was used for comparisons of three or more groups. p < 0.05 was considered to indicate a significant difference.