The human hepatocellular carcinoma cell lines Huh7, SMMC-7721, HepG2 and normal human liver cell lines LO2 were kindly provided by Dr Tongchuan He (University of Chicago Medical Center), and bone marrow stromal cell lines HS-5 were purchased from American Type Culture Collection (ATCC, Manassas, VA). Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM, HyClone, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, USA) and 100 U/ml streptomycin/penicillin at 37°C in 5% CO₂. The primary antibodies used in this study were: rabbit anti-Akt monoclonal antibody, rabbit anti-phosphor-Akt monoclonal antibody, rabbit anti-ERK1/2 monoclonal antibody and rabbit anti-phosphor-ERK1/2 monoclonal antibody and were obtained from Cell Signaling Technology. Rabbit anti-MMP-2 polyclonal antibody was purchased from Immunoway (Immunoway, USA). Mouse anti-β-actin monoclonal antibody was purchased from Santa Cruz Biothechnology. Secondary antibodies included HRP-conjugated goat anti-mouse IgG antibody and anti-rabbit IgG antibody were purchased from Zhongshan Goldenbridge Biotechnology. Specific inhibitors of PI3K (LY294002) and ERK1/2 (PD98059) were obtained from Cell Signaling Technology. CCL5 neutralization antibody was purchased from Peprotech.

HS-5 cells were plated in a 100 mm² culture dish at a density of 80%, for 6 h, for cells adherence, culture medium was changed to serum-free DMEM (SF DMEM). The culture supernatant was harvested every day for 3 days and then all medium were pooled, and sterile filtered. The HS-5-CM was stored at −20°C.

The co-culture experiment was set up in duplicates using 6-well Transwell inserts (Millipore, USA) with 0.4 μm pore size. In Transwell chamber, HS-5 cells were plated at the density of 10⁵ cells/well in 1 ml, whereas Huh7 or SMMC-7721 or LO2 cells were plated in the 6-well plates at the density of 2×10⁵ cells/well in 2 ml. Cells were left to adhere for 6 h before being put together in the SF DMEM for 3 days, and cells alone were used as control.

MTT [3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide] assay was used to detect cell viability. A total of 2×10³ cells/well were cultured in 96-well plates and treated with fresh DMEM, 50% HS-5-CM (CM:DMEM=1:1) or 100% HS-5-CM containing 1% FBS for 24, 48 and 72 h. After the indicated hours of culture, the MTT reagent (Sigma, St. Louis, MO, USA) was added (20 μl/well), and incubated for 4 h at 37°C. Dimethyl sulfoxide was added to dissolve the formazan product for 10 min at room temperature. Finally, the absorbance was measured daily for the next three days at 492 nm using a microplate reader. Each group was done in sextuplicate, and the overall experiment was repeated three times.

Cell cycle and cell apoptosis analysis were assessed by flow cytometry. Cells were divided into 3 groups: i) control group, Huh7 cells; ii) co-culture group, Huh7 cells were co-cultured with HS-5 cells; and iii) HS-5-CM group, Huh7 cells were treated with HS-5-CM. In brief, cells were seeded into 6-well plates at a density of 2×10⁵ cells/well. Every experiment was repeated twice in each group. Cells were harvested, re-suspended in 1 ml cold PBS. Samples were fixed with pre-cold 75% enthanol for 1 h at 4°C by cell cycle assay, and other samples were added using propidium iodide (PI) and FITC-Annexin V for cell apoptosis analysis according to the manufacturer’s protocol (Invitrogen, USA). Respectively, data were analyzed using FACsorter (Becton Dickinson, CA, USA).

For wound-healing assay, cells were grown to 95% confluence and the monolayer was scratched with a pipette tip, and then culture medium was replaced with fresh DMEM or 50% HS-5-CM or 100% HS-5-CM containing 1% FBS. Cells that migrated into the scratched area were compared using bright field microscopy at 48 h. Cell migration and invasion assay were performed by 24-well Transwell chambers (8 μm pore size, Millipore) without or with ECM gel (Sigma). Briefly, cells (5×10⁴/400 μl) in SF DMEM were seeded onto the upper chambers, and fresh DMEM or 50% HS-5-CM or 100% HS-5-CM (600 μl/well) containing 20% FBS added to the lower chambers. After 24 h, cells that migrated to the underside of the filter were fixed with methanol and stained with hematoxylin and eosin (H&E), and counted by brightfield microscopy. All the experiments were repeated three times.

The in vivo experiments were performed in accordance with the guidelines established by the Animal Care and Use Committee, of Chongqing University Laboratory Animal Research. The 4-6-week old male NOD/SCID mice were randomly divided into 2 groups (n=4/group), respectively, and were injected subcutaneously with a mixture of equal number of HS-5 cells (3×10⁶) and Huh7 cells (3×10⁶) or with Huh7 cells (6×10⁶) alone. Tumor volume was measured every 10 days with vernier calipers, and calculated in mm³ as ab²/2, where a and b represent the largest and smallest tumor diameters, respectively. The mice were sacrificed after 70 days, and tumor tissues, livers and lungs were collected and stained using H&E in order to examine the histopathology.

Huh7 cells were co-cultured with HS-5 cells or alone in SF DMEM for 3 days, total RNA was isolated using TRIzol Reagents (Invitrogen) according to the manufacturer’s instructions. Total RNA (2 μg) was used to generate cDNA templates for reverse transcriptase-PCR. Touchdown PCR analysis determined the gene expression levels and was performed by using the following program: 95°C × 2 min for one cycle, 10 cycles at 92°C × 20 sec, 65°C × 30 sec, and 70°C × 40 sec with a decrease of one degree per cycle, and 25 cycles at 92°C × 20 sec, 55°C × 30 sec, and 70°C × 40 sec. 70°C × 5 min for one cycle. Real-time PCR was run in the Rotor-Gene 6000 Real-Time PCR machine (Corbett Research, Australia) using SYBR Premix Ex Taq (Takara) with the following protocol: initial activation of HotStar Taq DNA polymerase at 95°C for 10 min, then 45 cycles of 95°C for 5 sec and 60°C for 20 sec. GAPDH was used as an internal control. The primer efficiency was confirmed to be high (>90%) and comparable (Table I). Data were analyzed according to the 2^−ΔΔCt method. The expression levels of mRNA were normalized to GAPDH.

To determine quantification of the levels of CCL5 secreted in the supernatants from co-culture or non co-culture, the culture medium was collected and detected by using ELISA Kits (RayBiotech, ELH-RANTES-001), according to the manufacturer’s protocol. Briefly, samples were added to the coated wells with CCL5 mAb of 96-well plates and incubated for 2.5 h at room temperature, washed four times and then incubated with an HRP-linked streptavidin solution for 30 min at room temperature in the dark. The absorbance was then measured at 450 nm on the microplate reader (Sunrise Remote, Switzerland) immediately. All the experiments were repeated for three times.

Equal amount of protein from Huh7 cells co-cultured with HS-5 cells or alone were mixed with SDS buffer, and were added onto a 5% (w/v) stacking polyacrylamide gel and separated on a 10% (w/v) polyacrylamide gel containing 1.0 mg/ml porcine gelatin (Sigma) for the detection of MMP-2 and MMP-9 activity. After electrophoresis, gels were washed twice for 40 min in 2.5% Triton X-100 (Sigma) to remove SDS, and incubated for 42 h at 37°C in 50 mM Tris-HCl (pH 7.6) containing 150 mM NaCl, 10 mM CaCl₂ and 0.02% Brij-35, and gels were stained for 3 h in 45% methanol/10% acetic acid containing 0.5% Coomassie brilliant Blue R-250 (w/v). Finally, destained with a solution containing 5% acetic acid until clear bands of gelatinolysis appeared on a dark background, proteolytic activity was detected as clear bands on a blue backgroup of the Coomassie Blue staining gel.

Western blot was done using standard protocols. In brief, Huh7 cells co-cultured or non co-cultured were washed three times with cold PBS and lysed in RIPA lysis buffer, and cell lysate was denatured after boiling. The protein concentration was measured according to NanoDrop100 UV-Vis Spectrophotometer (Thermo Scientific, USA). Equal amount of proteins were loaded onto 10% SDS-PAGE gels, and transferred subsequently onto PVDF membranes. After blocking with 5% BSA (Bovine Serum Albumin, Solarbio) in TBST, the membranes were probed with the primary antibody, and followed by incubation with a secondary antibody conjugating with horseradish peroxidase. Protein levels were quantified with SuperSignal West Pico Chemiluminescent Substrate kit.

Results are presented as mean ± standard deviation (SD). Comparisons were made using independent sample Student’s t-test in GraphPad prism 5. Statistical significance is indicated as p<0.05.

The effects of HS-5 cells on Huh7 cell viability in vitro were evaluated by MTT assay. It revealed that Huh7 cell viability treated with HS-5-CM increased obviously and time-dependently compared with control (Fig. 1). To further investigate the mechanisms of HS-5 cell-induced proliferation of Huh7 cells, flow cytometry was performed to measure the cell cycle. In order to study how HS-5 cells affect Huh7 cell growth in tumor microenvironment, we adopted two ways to culture Huh7 cells: co-cultured with HS-5 cells, and treatment with HS-5-CM, subsequent experiments were done using 100% HS-5-CM. Respectively, Huh7 cells in co-culture group (37.85±1.52%) and in HS-5-CM group (29.56±2.03%) were at S phase of cell cycle at day 3, which was significantly higher than the control (8.96±2.34%) (Fig. 2A). These two ways had different degrees of increase in S phase, but the data demonstrated that HS-5 cells could promote Huh7 cell proliferation.

The proportion of Huh7 cells undergoing apoptosis was determined by flow cytometry analysis. After 3 days, the data showed the apoptosis ratio (early + late) of Huh7 cells in co-culture group was approximately 6.50%, and it was significantly decreased compared to the control group, although HS-5-CM was not distinct in decreasing the apoptotic percentage of Huh7 cells, the small change implied that HS-5 cells could suppress Huh7 cell apoptosis (Fig. 2B).

As cell migration and invasion abilities are important determinants in the progression of HCC metastasis, the potential to migrate was tested in the commonly used wound-healing assay. After cultured with HS-5-CM for 48 h, the motility of Huh7 cells was significantly accelerated compared with control, especially in 100% HS-5-CM (Fig. 3A). Transwell assay was carried out to verify the migration potency. In line with the wound-healing experiment, it demonstrated that 100% HS-5-CM prominently increased the migration potency of Huh7 cells compared to the control (Fig. 3B). Applying Transwell assay to test the invasion property of Huh7 cells for 24 h, respectively, the number of transmembrane cells in 50% HS-5-CM group and 100% HS-5-CM group was remarkably increased compared to the control group (Fig. 3C).

A critical experiment for determining whether HS-5 cells affect tumor growth of HCC is the in vivo tumorigenicity assay. Therefore, we sought to validate our in vitro findings by using an in vivo model. It found that HS-5 cells accelerated the growth of Huh7 tumors, up to 20 days, there was no significant difference in the two groups, while the tumor volumes became palpable after 20 days, the Huh7 group grew from 150 to 2,522.4 mm³ and the co-culture group grew from 200 to 3,321.8 mm³ (Fig. 4A and B). These results were consistent with the in vitro observation that HS-5 cells markedly promoted Huh7 cells growth. H&E staining showed no variance in heterogeneity between the two groups (Fig. 4C). Tumor cells were not found in the liver and lung tissues between the two groups (data not shown).

The molecular mechanisms of these pro-metastatic impacts in tumor microenvironment are still unclear. To investigate whether it is mediated by soluble factors secreted from HS-5 cells in the capacity of migration and invasion, we identified several chemokines by RT-PCR using RNA prepared from HS-5 cells co-cultured with Huh7 cells or alone at day 3. It was found that CXCL12, CX3CL1, CXCL8, CCL5 mRNA levels were upregulated in HS-5 cells after co-culture, among which CCL5 showed the highest level of increase (Fig. 5A). We also examined these chemokine special receptor mRNAs in Huh7 cells. RT-PCR assay showed that CXCR4 and CCR5 mRNA expression was increased, and CX3CR1 mRNA was decreased in Huh7 cells co-cultured with HS-5 cells compared with Huh7 cells alone at day 3 (Fig. 5B).

To confirm the change on CCL5 expression by qRT-PCR and ELISA assay. qRT-PCR analysis demonstrated that CCL5 mRNA expression had an approximate 5-fold increase in HS-5 cells co-cultured with Huh7 cells compared with HS-5 cells alone at day 3 (Fig. 5C). In accordance with CCL5 mRNA level, ELISA assay revealed that co-culture specifically increased CCL5 protein level more than 3-fold compared with the secretion from Huh7 cells or HS-5 cells (Fig. 5D).

ELISA assay also verified that the change of CCL5 upregulation was not an accidental event with SMMC-7721 or HepG2 or LO2 cells. It found that CCL5 protein levels were increased in HS-5 cells co-cultured with SMMC-7721 or HepG2 cells in different degrees, but not with normal liver cells LO2 (Fig. 5E). The aforementioned data illustrated that interaction with HCC and HS-5 cells could increase CCL5 secretion in HS-5 cells.

Whether HS-5 cells can modulate Huh7 cells progression via CCL5 secretion remains unreported. We took a loss-of-function approach to confirm its impact on Huh7 cells, utilizing anti-CCL5 neutralizing antibody to HS-5-CM which was cultured in Huh7 cells for 24 h. We tested the migration and invasion abilities by Transwell assay, and it illustrated that anti-CCL5 neutralizing antibody caused a decrease of 27.47% in HS-5-CM-induced migration activity of Huh7 cells compared with the control (Fig. 6A). Consistently, Huh7 cell invasion ability was inhibited by 33.06% compared to the control (Fig. 6B). However, not as imagined, CCL5 downregulation in HS-5-CM did not overtly change proliferation and apoptosis of Huh7 cells (Fig. 6C and D). It was thus possible that CCL5 from HS-5-CM did not play a leading role in Huh7 cell proliferation and apoptosis impact, but in migration and invasion potency.

The underlying mechanisms of the interaction with MSCs and HCC within HCC micro-environment remain an enigma, so we concentrated on the mechanisms of HS-5 cells inducing migration and invasion of Huh7 cells. It has been reported that stimulation of MSCs with certain chemokines causes the activation of Akt kinase and ERK1/2 MAP kinase (31). Therefore, we examined the potential changes of Huh7 cells in the presence of HS-5 cells further. The transwell co-culture with HS-5 cells for 3 days, demonstrated by western blot analysis, revealed that Huh7 cells had enhanced expression of p-Akt (Ser473) (Fig. 7A), but not p-ERK1/2 (Fig. 7B). The aforementioned data showed that CCL5 expression was upregulated in HS-5 cells after co-culture (Fig. 5C and D). Previous research showed that CCL5 activated Gαi-PI3K-Akt and Gαi-MEK-ERK signaling pathways (32). So it was postulated that CCL5 is associated with the activation of p-Akt (Ser473) and p-ERK1/2 in co-culture system. We applied anti-CCL5 antibody for 24 h to the co-culture system to evaluate the hypothesis, the data showed the activation of p-Akt (Ser473) was reduced by adding anti-CCL5 neutralizing antibody to Huh7 cells co-cultured with HS-5 cells (Fig. 7A), however, there was no significant change in p-ERK1/2 (Fig. 7B). We next focused on CCL5 expression in response to p-Akt (Ser473) and p-ERK1/2 signaling pathways. Before we investigated the roles of p-Akt (Ser473) and p-ERK1/2 signaling pathways, western blot analysis demonstrated PI3K and ERK1/2 specific inhibitors which, respectively, were LY294002 (20 μM) and PD98059 (10 μM) could decrease the expression of p-Akt (Ser473) and p-ERK1/2 (Fig. 7A and B). Transwell assay illustrated that the migration and invasion of Huh7 cells induced by HS-5-CM were partly reversed by treatment with LY294002 but not by PD98059 (Fig. 7C and D). In addition, ELISA assay demonstrated that LY294002 reduced the secretion of CCL5 in co-culture system (Fig. 7E). Thus, our results illustrated that CCL5 derived from HS-5 cells had a pivotal role in migration and invasion of Huh7 cells via PI3K-Akt signaling pathway.

Cell movement is involved in the proteolytic activity of MMPs, which regulate the dynamic ECM (extracellular matrix)-cell and cell-cell interaction during migration (33). High expression of MMPs is related to HCC progression (34,35). We investigated whether CCL5 regulated the levels of MMP-2, MMP-7 and MMP-9 in Huh7 cells co-cultured with HS-5 cells. RT-PCR analysis showed that MMP-2 expression was up-regulated in Huh7 cells co-cultured with HS-5 cells at day 3, extraordinary, MMP-2 mRNA was downregulated through the addition of anti-CCL5 neutralizing antibody. Nonetheless, the expression of MMP-7 and MMP-9 mRNAs remained significantly unchanged (Fig. 8A). To further characterize the effect of CCL5 in MMP expression of Huh7 cells, we applied gelatin zymography analysis to determine expression and enzymatic activities of MMP-2 and MMP-9. The results revealed that treatment of anti-CCL5 neutralizing antibody pointed to a decrease in MMP-2 upregulation of Huh7 cells co-cultured with HS-5 cells, whereas MMP-9 was not detected (Fig. 8B). Consistently, western blot assay confirmed that anti-CCL5 neutralizing antibody could reduce MMP-2 protein upregulation mediated by HS-5 cells in Huh7 cells (Fig. 8C). The data indicated that CCL5 downregulation from HS-5 cells decreased MMP-2 expression in Huh7 cells.