Activated Hepatic Stellate Cells Promote Angiogenesis Via IL-8 In Hepatocellular Carcinoma

IntroductionHepatocellular carcinoma(HCC) is one of the most common malignant solid tumors and the prevalence of HCC in China preced es any other countries in the world. Although the treatment for HC C has progressed to some extent with the development of medicine,the tumor freeand total survival rates of HCC are still unsatisfying.Statistically, 60%-70% patients still suffer from reoccurrence and me tastasis within five years even after radical dissection.Therefore,it is urgent to elucidate factors associated with the growth,reoccurrence and metastasis of HCCand the underlying regulatorymechanism.Previous studies concerning the growth, reoccurrence and metastasis of HCC mainly focus on hepatoma cells. However, accumulating evidence indicates that the occurrence and development don’t solely depend on hepatoma cells,but also on themicroenvironment of HCC.The occurrence and development ofhepatocellular carci-noma(HCC) take place in a complicatedmicroenvironment,where the angiogenesis, reoccurrence andmetastasis of HCC are determined by the interaction betweenhepatoma cells and the surrounding stroma. The microenvironmentof hepatocellular carcinoma comprisesactivated hepatic stellate cells(a-HSCs), endothelial cells, epithelial cells, immune cells, variousmetabolites and extracellular matrix. HSCs are the predominant ant cell type in the stroma of HCC, Chemokines and cytokines secretedby HSCs,through both autocrineand paracrinemodes,play a key rolein the angiogenesis, growth and metastasis of HCC,influencing the prognosis of HCC directly.HSCs in two cellular states, dormant or activated, are the predominant ant cell type in the stroma of HCC. Under the stimulation of inflammation and tumor, a-HSCs are converted from dormant to activated state. Not only can a-HSCs promote the invasiveness and growth of tumor, but also induced endothelial cells in large quantity to migrate into the tissue of HCC, thus promoting the angiogenesis of HCC. However,these new vessels are prominently abnormal, usually arterialized, capillary-likevessels. It eventually lead to the speeded neovascularization of HCC, thus facilitating the growth and metastasis of HCC.Currently it is believed that neovascularizationis an essential part of the endless and incontrollable growth of malignant tumor. Neovascularization in HCC is essential to the growth, invasion, reoccurrence and metastasis of hepatoma. Recently the formation and inhibition of neovascularization has become the emphasis of research. It has been demonstrated in studies on cirrhosis-induced portal hypertension that activated hepatic stellate cells(a-HSCs) promote angiogenesis in liver and cause the remodeling of sinusoids,thus leading to portal hypertension. In studies about metastatic melanoma, angiogenesis first occurs where a-HSCsgather;furthermore,the density of angiogenesispositively correlates the number of a-HSCs. These results suggest that angiogenesis participates in the angiogenesis of HCC.However, currently the mechanism on how a-HSCs influence the angiogenesis in HCC is still not fully understood. For instance,what cytokines are secreted by a-HSCs? How to promote the angiogenesis in HCC by influencing hepatoma cells? Aiming at resolving these problems, we plan to further investigate theproangiogenic effects of hepatoma cells by establishing the modelwhere hepatoma cells are treated with supernatants from a-HSCs. ①Multiplex bead-based enzyme-linked immunosorbent assay and enzyme-linked immunesorbentassay(ELISA) were used to test whethera-HSCs secrete inflammatory cytokines. ②Chick embryo chorioallantoic membrane(CAM) assay and capillary tube formation assay wereused to study the proangiogenic effect of IL-8.③The role ofp-STAT3 signaling pathway in the regulation of angiogenesis by IL-8 was investigated with methods like immunoblot. The present studywill give us a bett er understanding of the microvascular formation of HCC and a new insight into the treatment for HCC.Materials and methodsPatients and Clinical Tissue CollectionAll tumor samples pathologically confirmed for hepato- cellular carcinoma were obtained from 20 patients undergoing hepatectomy between 2013-2014 from the Department of Hepatobiliary Surgery of the Third Affiliated Hospital of Sun Yat-sen University in Guang zhou. These patients had signed informed consent and had received no previous local or systemic treatment before operation. The experiment was conducted in strict accordance with a study design approved by the clinical Research Ethics Committee at the Third Affiliated Hospital of Sun Yat-sen University in Guangzhou.Isolation and culture of hepatic stellate cellsThe method of separation of hepatic stellate cells from fresh tissues have been described in previous studies.In order to minimize culture stress and clonal selection, hepatic stellate cells passagedfor up to 4–10 passage doublings were used for subsequent experiments.Isolation of Human Umbilical Vein Endothelial Cells(HUVECs) and Capillary Tube Formation AssayHUVECs were isolated from fresh human umbilical cords with collagenase I. After dissociation, the cells were collected and cultured on 25cm2 culture flasks in medium 199(m199, Life Technologies) supplemented with 20% fetal bovine serum(FBS, GIBCO, Aus-tralia),0.03mg/ml of endothelial cell growth supplement(ECGS Ups tate Biotechnology, Lake Placid, NY) and 0.1mg/ml of Heparin in a5% CO2 incubator at 37℃. Only primary HUVECs of 3–6 passages were used in this study to avoid age-dependent cellular modifications. Diluted Matrigel(Matrigel: M199=1:2)(R and D Systems) was added to a 96-well plate(50 ml/well) and left to polymerize for 2 h at 37℃. HUVECs were then added at a density of 3x104 cells per well onto the 96-well plate in different serum-free conditioned medium from hepatoma cells(Hep3B and Huh-7) or hepatoma cells(Hep3B and Huh-7) exposed to HSCs serum-free culture supe rnatant in the absence or presence of the IL-8 neutralizing antibody for 8h in a 5% CO2 incubator at 37℃. Blood vessel branch points were measured by phase microscopy in 3 random fields per well(10x).Cell CultureThe human hepatoma cells(Hep3B and Huh-7) were purchased from Chinese Academy of Sciences. All cell lines were cultured in DMEM(Dulbecco’s modified Eagle’s medium, Hyclone, Logan, UT) supplemented with 10%FBS in a 5% CO2 incubator at 37℃.Immunohistochemical stainingAll tumor specimens were fixed in formalin-free IHC zinc fixa-tive and then embedded in paraffin. These specimens were cut into 4 micron–thick paraffin sections and then fixed on the mircoslide. There are three sections for immunohistochemical analysis on each mircoslide. These slides were baked for 30 min at 55℃ in oven and were deparaffinized in xylene, hydrated with graded ethanol(100%, 95%, 80%, and 70%). After eliminating the endogenous peroxidase activity with 0.3% hydrogen peroxide and repairing antigen with citric acid(PH 6.0), these specimens were incubated overnight at 4℃ with a-smooth muscle actin(a-SMA) antibody(1:1000)(Abcam, Cambridge, MA), CD34 antibody(1:50)(e Bioscience, San Diego, CA) and IL-8 neutralizing antibody(1:400)( R and D Systems). After incubation for 30 min at 37℃ with a secondary antibody, coloration were performed before washing with Phosphate Buffer Sol-ution(PBS) four times. All mircoslides were counter stained with hematoxylin and observed under the microscope.Immunofluorescence assayA-HSCs were cultured in 24-well flat-bottom plates, when a-HSCs were approximately 40-70% confluent, immunofluorescence assay was executed as previously described. In short, after fixation with 100% cold carbinol(10 minutes), a-HSCs were incubated with rabbit anti-human monoclonal antibodies-a-SMA, vimentin and immunoglobulin(Ig G)(Abcam Cambridge, MA) as controls in Tris-buff ered saline(p H 7.4) for at 4℃ overnight. The cells were then rinse d and prewetted with PBS and incubated at 37℃ for 1 hour with AF488-conjugated donkey anti-rabbit Ig G(Molecular Probes, Carlsba d, CA), the cell nuclei were counterstained with 40-6-diamidino-2-p henylindole(Sigma Aldrich, St. Louis, MO). These images were ass essed by a fluorescence microscope(LEICA DMI 4000 B, Germany) at a wavelength of 488 nm and analyzed with Leica Application suite software(version 4.0).Enzyme-Linked Immunesorbent Assay(ELISA)Hepatoma cells(Hep3B and Huh-7) were stimulated with a-HSCs conditioned medium(HSC-CM)(CM : HSC-CM =1:1) with orwithout IL-8 neutralizing antibody for 24 h, conditioned medium of the above mentioned cells and untreated hepatoma(Hep3B and Hu h-7) cells were collected and centrifuged to remove cellular debris. The levels of IL-8 and Vascular endothelial growth factor –A(VEGF-A) were quantified by ELISA kits(R and D Systems) as previously described.Chick embryo chorioallantoic membrane(CAM) assayWhite leghorn chicken eggs on the fifth day of fertilization were purchased from Animal Husbandry Institute of South China Agricultural University. The air chamber of these eggs was windowed after disinfection by 75% alcohol. The shell membrane was then removed carefully and a 50 ul droplet was pipetted onto the CAM. These windows were sealed with sterile ventilate cellotape and incubation continued without rotation at 80% relative humidity and 37.8°C in the incubator. After 48 h, the results were observed under a ZEISS Lumar.V12 stereomicroscope.Multiplex bead-based enzyme-linked immunosorbent assayAccording to the manufacturer’s instructions,supernatants were generated by seeding 5x104 cells per well into 48-well plates. The Multiplex bead-based enzyme-linked immunosorbent assay analysisof the conditioned supernatants was performed using the Human 38-plex antibody bead kit, the Human11-plex antibody bead kit and the Human 23-plex antibody bead kit(Millipore, Billerica, MA, US A). 200μL Assay Buffer was added to each well of the test plate, sealed, mixed at room temperature for 10 minutes and aspirated with a vacuum pump. 25μL standard substance, reference substance,substrate solvent was added to the corresponding well subsequently. Next,25μL testing beads were added to each well and incubated for1 h at room temperature. The mixture was then aspirated slowly and each well was rinsed twice with 200μL Wash Buffer. 25μL testing antibody was added to each well and incubated for 1h. Then 25μL Streptavidin-Phycoerythrin was added and incubated for another 30 min.The plate was washed twice with 200μL Wash Buffer followed by further reaction for 30 min in 150μL sheath fluid. Results were analyzed using a Luminex plate reader and Milliplex analyst software(Luminex 200 System).Western BlotWhen hepatoma cells(Hep3B and Huh-7) grew to 70%- 80% confluence, cells were stimulated with HSC-CM(CM : HSC-CM =1:1) with or without IL-8 neutralizing antibody for 24 h in 6-well plates. Total cell protein was extracted using Triton Lyses buffer af-ter the wells were washed at least twice. Twenty microlitresamples were loaded onto 10% or 15% SDS-acrylamide gels and then transferred to 0.20μm nitrocellulose membranes. Immunoblotted with an Ab against glyceraldehyde-phosphate dehydrogenase(GAPDH), the protein expression of IL-8 and other targeted markers was detected with an ECL kit. Antibodies for western blot were as follows: rabbit m Ab for nuclear factor κB(NF-κB), rabbit m Ab for phosphonu clear factor κB(p-NF-κB), mouse m Ab for Signal transducer and activator of transcription 3(STAT-3), rabbit m Ab for phospho-Signal transducer and activator of transcription 3(p-STAT-3), HIF-1, GAPDH and β-actin(Cell Signaling Technology, CST, Beverly, MA).Real-Time Quantitative polymerase chain reaction(q PCR) AnalysisTotal RNA of cultured Hepatoma(Hep3B and Huh-7) cells was extracted using Tri Pure Isolation Reagent(Roche Applied Science, Germany) according to the manufacturer’s instructions.The sequences of primers about angiogenic factors used for q PCR are described in the Table 1. The q PCR analysis of m RNA was executed accord ing to the manufacturer’s instructions. All reactions were performed in triplicate.Statistical analysisThese results are expressed as the mean±standard error of the mean(SEM). the differences between groups were analyzed using Student’s t test when only two groups were compared or by ANOVA(one-way analysis of variance) when more than two groups were compared. All statistical tests were two-sided. All experiments were carried out from at least three independent experiments. p <0.05 were considered statistically significant.Results1. A-HSCs successfully isolated and successively cultured from the stroma of HCC highly expressedα-SMA and vimentin.2. Among the inflammatory chemokines that promoted angiogenesis,GRO(GROα、GROβ、GROγ)、CXCL5、CXCL6、CXCL7 and IL-8, the concentration of IL-8 was evidently higher than that of other chemokines.3. In hepatoma tissues, chemokine IL-8 was mainly expressed in th e stroma, and slightly in the nest.4. Compared to conditioned medium(CM), supernatant collected from hepatoma cells treated with 50% HSC-CM could significantly promote angiogenesis. However, IL-8 neutralizing antibody markedly reduced such proangiogenic effects.5. IL-8 neutralizing antibody significantly inhibited the secretion ofpro-angiogenic factors and their m RNA expressioninhepatoma cells treated with 50% HSC-CM.6. IL-8 neutralizing antibody significantly inhibited the p-STAT3 signaling pathway within hepatoma cells, thus inhibiting neovascularization. However, the STAT3、p-NF-κB、NF-κB、HIF-1α signaling pathway within HCC remained unchanged.ConclusionIn the microenvironment of HCC, inflammatory chemokine IL-8 in large quantity was secreted by a-HSCs, combined the receptor on the surface of HCC and promoted the secretion of angiogenic factors through the p-STAT3 signaling pathway in the hepatoma microenvironment, thus leading to increased neovascularization on the invading edge that was close to the tumor stroma and promoting th e occurrence, development, infiltration and metastasis of HCC. The present study gives us a better understanding of the microvascular formation of HCC and a new insight into the treatment for HCC.