Conceived and designed the experiments: IG MM SW MA FG MZ JS UML AAS. Performed the experiments: IG MM SW JS. Analyzed the data: IG MM MA SW MH NGC JS UML AAS. Contributed reagents/materials/analysis tools: MB QZ YAY. Wrote the paper: IG UML AAS.
¶ These authors also contributed equally to this work.
The authors have read the journal's policy and have the following conflicts: IG QZ YAY NGC JS and AAS are employees and shareholders of Genelux Corporation. However, no direct financial support was provided by Genelux for the completion of this study. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials, as detailed online in the guide for authors. All of the other authors declare that they have no conflict of interest.
Virotherapy using oncolytic vaccinia virus strains is one of the most promising new strategies for cancer therapy. In this study, we analyzed for the first time the therapeutic efficacy of the oncolytic vaccinia virus GLV-1h68 in two human hepatocellular carcinoma cell lines HuH7 and PLC/PRF/5 (PLC) in cell culture and in tumor xenograft models. By viral proliferation assays and cell survival tests, we demonstrated that GLV-1h68 efficiently colonized, replicated in, and did lyse these cancer cells in culture. Experiments with HuH7 and PLC xenografts have revealed that a single intravenous injection (i.v.) of mice with GLV-1h68 resulted in a significant reduction of primary tumor sizes compared to uninjected controls. In addition, replication of GLV-1h68 in tumor cells led to strong inflammatory and oncolytic effects resulting in intense infiltration of MHC class II-positive cells like neutrophils, macrophages, B cells and dendritic cells and in up-regulation of 13 pro-inflammatory cytokines. Furthermore, GLV-1h68 infection of PLC tumors inhibited the formation of hemorrhagic structures which occur naturally in PLC tumors. Interestingly, we found a strongly reduced vascular density in infected PLC tumors only, but not in the non-hemorrhagic HuH7 tumor model. These data demonstrate that the GLV-1h68 vaccinia virus may have an enormous potential for treatment of human hepatocellular carcinoma in man.
Hepatocellular carcinoma (HCC) is one of the most common malignancies worldwide
Here, we have investigated the therapeutic potential of the oncolytic vaccinia virus GLV-1h68 against HCC in preclinical studies. GLV-1h68 was derived from vaccinia virus Lister strain (LIVP) that contains an inactive thymidine kinase (tk) gene and shows inherently more tumor-selective replication than vaccinia virus WR strain
We and others have already demonstrated tumor selectivity and efficacy of GLV-1h68 in many different tumor xenograft models, including human breast cancer
Here, we describe that GLV-1h68 was able to infect, replicate in, and lyse human hepatocellular carcinoma cell lines HuH7 and PLC/PRF/5.
We also found that a single intravenous injection of GLV-1h68 into mice with subcutaneously grown hepatocellular carcinoma xenografts dramatically reduced tumor growth. Lastly, the oncolytic and immunological effects of GLV-1h68 in HCC tumors were analyzed by fluorescence imaging, immunohistochemistry, flow cytometry (FACS) and immune-related protein antigen profiling.
All animal experiments were approved by the government of Unterfranken, Germany, and conducted according to the German animal protection guidelines (permit number: 55.2–2531.01-17/08).
African green monkey kidney fibroblasts (CV-1, American Type Culture Collection, ATCC-No. CCL-70) and two human hepatocellular carcinoma cell lines HuH7 (ATCC CCL-185) and PLC/PRF/5 (PLC; ATCC CRL 8024) were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with antibiotics (100 units/ml penicillin G, 100 units/ml streptomycin) and 10% fetal bovine serum (FBS; Invitrogen GmbH, Karlsruhe, Germany) at 37°C under 5% CO2.
GLV-1h68 is a genetically stable oncolytic virus strain designed to locate, enter, colonize and destroy cancer cells without harming healthy tissues or organs
HuH7 and PLC cells were seeded onto 24-well plates (Nunc, Wiesbaden, Germany). After 24 h in culture, cells were infected with GLV-1h68 using multiplicities of infection (MOI) of 0.1 and 1. Cells were incubated at 37°C for 1 h, then the infection medium was removed and the cells were incubated in fresh growth medium. The amount of viable cells after infection with GLV-1h68 was measured using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Sigma, Taufkirchen, Germany). At 24, 48, 72, or 96 h after infection of cells, medium was replaced by 0.5 ml MTT solution at a concentration of 2.5 mg/ml MTT dissolved in RPMI 1640 without phenol red and incubated at 37C° for 2 h in a 5% CO2 atmosphere. After removal of the MTT solution, the color reaction was stopped by adding 1 N HCl diluted in isopropanol. The optical density was then measured at a wavelength of 570 nm. Uninfected cells were used as reference and were considered as 100% viable. The amount of viable cells after infection with GLV-1h68 was measured in triplicates.
HuH7 and PLC cells grown in 24-well plates were infected with GLV-1h68 at an MOI of 0.1. After incubation at 37°C for 1 h with gentle agitation every 20 min, the infection medium was removed and replaced by fresh growth medium. Supernatants were collected from virally treated cells at 1, 6, 12, 24, 48, 72 or 96 h post-infection. Serial dilutions of supernatants were titrated by standard plaque assays on CV-1 cells. All samples were measured in triplicates.
The GFP signals of virus-infected cells were analyzed with a fluorescence microscope (Leica DM IRB; Wetzlar, Germany). Images were captured with an electronic camera and were processed using META-MORPH (Universal Imaging; Downingtown, PA, USA) and Photoshop 7.0 (Adobe Systems, Mountain View, CA, USA).
Tumors were generated by implanting hepatoma cells HuH7 or PLC (5×106 cells in 100 µl of PBS) subcutaneously on the right flank above the hind leg of 6- to 8-week-old male or female nude mice (NCI/Hsd/Athymic Nude-
The statistical significance of the data was calculated by two-way analysis of variance (ANOVA) with Bonferroni comparison post-test (GraphPad Prism software, San Diego, USA). The post-test was only performed when ANOVA revealed significance. Results are displayed as means ± s.d.
For histological studies, tumors were excised and snap-frozen in liquid N2, followed by fixation in 4% paraformaldehyde/PBS at pH 7.4 for 16 h at 4°C. Tissue sectioning was performed as described by Weibel et al.
Endothelial cells were labeled with monoclonal rat anti-mouse CD31 antibody (BD Pharmingen, San Diego, CA) or hamster anti-mouse CD31 antibody (Chemicon, International, Temecula, CA). Immune cells were labeled using rat anti-mouse MHCII antibody detecting a polymorphic determinant present on B cells, monocytes, macrophages and dendritic cells (eBioscience, San Diego, CA). The Cy3- or Cy5-conjugated secondary antibodies (donkey) were obtained from Jackson ImmunoResearch (West Grove, PA).
The fluorescence-labeled preparations were examined using the MZ16 FA Stereo-Fluorescence microscope (Leica) equipped with the digital DC500 CCD camera and the Leica IM1000 4.0 software (1300×1030 pixel RGB-color images) as well as the Leica TCS SP2 AOBS confocal laser microscope equipped with an argon, helium-neon and UV laser and the LCS 2.16 software (1024×1024 pixel RGB-color images). Digital images were processed with Photoshop 7.0 (Adobe Systems, Mountain View, CA, USA) and merged to yield overlay images.
The vascular density was determined in microscopic images (x10 objective, x10 ocular, tissue region 1445 µm by 1445 µm) of CD31-labeled tumor sections captured with identical settings using the Leica TCS SP2 AOBS confocal laser microscope. All images were decorated with eight horizontal lines at identical positions using Photoshop 7.0 and all vessels which intersected these lines were counted to yield the vascular density. The vascular density was calculated for six images per group and presented as mean values ± standard deviations (s.d.).
For preparation of tumor lysates, at 10 days after virus treatment, three mice from each group were sacrificed. Tumors were removed, resuspended in 9 volumes (W/V) lysis buffer [50 mM Tris-HCl (pH 7.4), 2 mM EDTA (pH 7.4), 2 mM PMSF and Complete Mini protease inhibitors (Roche, Mannheim, Germany)] and lysed using FastPrep FP120 Cell Disruptor (BIO 101, Qbiogene, Germany) at a speed of 6 for 20 s (three times). Samples were then centrifuged at 20,000 g at 4°C for 5 min and the supernatants were analyzed for mouse immune-related protein antigen profiling by Multi-Analyte Profiles (mouse MAPs; Rules Based Medicine, Austin, USA) using antibody linked beads. Results were normalized based on total protein concentration.
For flow cytometric analysis, three or four mice from each group were sacrificed by CO2 inhalation and the tumors were removed. The tumor tissues were minced and incubated individually in 10.000 CDU/ml Collagenase I (Sigma, Steinheim, Germany), 32 mg/ml Dispase II (Roche Diagnostic, Mannheim, Germany) and 5 MU/ml DNase I (Calbiochem, Darmstadt, Germany) for 40 min at 37°C and then passed through a 70-µm nylon mesh filter (BD Biosciences, Erembodegem, Belgium). Cells were incubated at 4°C for 40 min in PBS with 2% FCS, in the presence of appropriate dilutions of labeled monoclonal antibodies: anti-mouse MHCII-PE (Clone M5, eBioscience, Frankfurt, Germany), anti-CD19-PE-Cy5.5 (Clone 6D5, Beckman Coulter, Krefeld, Germany), anti-F4/80-APC (Clone BM8, eBioscience), and anti-Ly6G-PE (Clone 1A8, BD Biosciences). Stained cells were subsequently analyzed, using an Accuri C6 Cytometer and FACS analysis software CFlow Version 1.0.227.4 (Accuri Cytometers, Inc. Ann Arbor, MI USA).
The replication efficiency of GLV-1h68 in HuH7 and PLC cells was analyzed as described in
Supernatants were collected from virus-infected cells at various time points (hours) post-infection (hpi). Viral titers were determined as pfu per well in triplicates. Averages plus standard deviation are plotted.
In order to test the ability of the GLV-1h68 virus to infect and lyse HuH7 and PLC cells we performed a cell viability assay (
Viable cells after infection with GLV-1h68 were determined by use of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Sigma, Taufkirchen, Germany). Mean values (n = 3) and standard deviations are shown as percentages of respective controls.
To confirm the efficient infection and replication of GLV-1h68 in hepatocellular carcinoma HuH7 and PLC cells, we followed the virus-mediated expression of the
These results indicated that GLV-1h68 was able to efficiently infect and kill both, HuH7 and PLC cells in cell culture.
To test the therapeutic efficacy of GLV-1h68 against human hepatocellular carcinoma
Groups of HuH7 (
In the animal studies previously described, we observed a phenotypic difference (switch) in the tumor appearance of virus-infected versus mock-treated PLC tumor-bearing mice by day 14 p.i. The control mice developed dark bluish, hemorrhagic tumors as shown in (
Pictures were taken at day 18 after injection with PBS (mock) (A, C) or GLV-1h68 (B, D).
Mice bearing tumors of HuH7 (A, B) or PLC (C, D) origin either were mock treated (A, C) or infected with GLV-1h68 (B, D). Tumor sections were labeled with an anti-MHCII antibody (red) and viral infection was indicated by GFP fluorescence (green). In addition, overlays of MHCII and GFP signals and transmission images (bright-field) are shown. Scale bars represent 1 mm.
These immunohistological data were also quantitatively analyzed and verified by flow cytometric analysis (FACS) of tumor single cell suspensions derived from infected and uninfected HuH7 and PLC tumors at 7dpi (
Tumor |
GLV-1h68/untreated Ratio (HuH7) | GLV-1h68/untreated Ratio (PLC) | uninfected HuH7/uninfected PLC |
|
|||
MHCII | 2.28 | 5.68 | 4.68 |
F4/80 | 1.01 | 3.34 | 2.79 |
CD19 | 1.56 | 5.63 | 6.41 |
Ly6G | 1.86 | 10.22 | 12.98 |
1.20% +/− 0.012% | 7.10% +/−0.044% |
Markers: MHCII-PE antibody detects a polymorphic determinant present on B cells, monocytes, macrophages and dendritic cells.
F4/80-APC antibody recognises the F4/80 antigen, that is expressed by a majority of mature macrophages and is the best marker for this population of cells.
CD19 is expressed on B cells and follicular dendritic cells.
Ly6G-PE, also known as Gr-1- antigen, is expressed on mouse neutrophils, predominantly granulocytes.
Single cell suspensions derived from infected and uninfected HuH7 and PLC tumors 7dpi (n = 4 for uninfected and n = 3 for virus-infected groups) were used for FACS characterization. Ratios greater than 1 indicate an increased accumulation of host immune cells.
GFP-positive cells of the infected tumors were presented as mean values (n = 3) +/− standard deviations in percentages.
The increased presence of host immune cells in tumor tissue before virus injection might be responsible for the different infection and therapeutic efficacies of GLV-1h68 in these two xenograft models.
In order to study the influence of the tumor microenvironment on the efficiency of cancer therapy, we analyzed and compared the protein profiles of infected and uninfected HuH7 and PLC tumors. Lysates of tumors were prepared and aliquots used for examination of the expression levels of immune-related proteins of mouse origin, as described in
Antigen | GLV-1h68 / untreated Ratio (HuH7) | GLV-1h68 / untreated Ratio (PLC) | Classification |
GM-CSF | 2.18 | 6.23 | granulocyte-macrophage colony-stimulating factor |
IFN-gamma | 1.29 | 1.24 | proinflammatory cytokine |
IL-6 | 5.67 | 11 | proinflammatory cytokine |
IL-12 (IL-12p70) | 3 | 4.27 | pleiotropic cytokine |
IL-18 | 1.2 | 3.7 | proinflammatory cytokine |
IP-10 (CXCL10) | 3.12 | 112.2 | interferon-gamma-induced protein |
MCP-1 (CCL2) | 6.55 | 40.9 | proinflammatory cytokine |
MCP-3 (CCL7) | 2.06 | 26.6 | proinflammatory cytokine |
MCP-5 (CCL12) | 18.17 | 40.6 | proinflammatory cytokine |
MPO | 0.9 | 70 | myeloperoxidase |
M-CSF-1 | 1.15 | 5.23 | proinflammatory cytokine |
MIP-1beta | 1.59 | 6.57 | proinflammatory cytokine |
MIP-2 (CXCL2) | 0.81 | 15.44 | proinflammatory chemokine |
TNF-alpha | 1.2 | 2.91 | proinflammatory cytokine |
MIP-1gamma (CCL9) | 0.03 | 0.0227 | macrophage inflammatory protein |
Factor VII | 1.045 | 0.047 | plays a role in coagulation cascade |
Fibrinogen | 0.812 | 0.126 | plays a role in coagulation cascade |
GST-alpha | 0.92 | 0.0071 | biomarker |
Haptoglobin | 1.632 | 0.17 | biomarker |
All ratios greater than 1 indicate an up-regulation of the protein expression, and all ratios less than 1 indicate down-regulation.
Thus, the drastic reduction of fibrinogen and factor VII in the course of application of GLV-1h68 could also be an explanation for the loss of the hemorrhagic phenotype in virus-treated PLC human hepatoma xenografts.
Lastly, we also compared the protein profiles of uninfected HuH7 and PLC derived tumors (
Antigen (least detectable dose |
HuH7 tumor tissue | PLC tumor tissue | Classification |
CD40 |
272 pg/ml | 109 pg/ml | a type I glycoprotein belonging to the TNF receptor superfamilyjavascript:void(0); |
Factor VII (FVII) |
198 ng/ml | 851 ng/ml | plays a role in coagulation cascade (increased FVII indicates blood vessel injury) |
Fibrinogen |
38 µg/ml | 300 µg/ml | plays a role in coagulation cascade |
GST-alpha |
6.3 ng/ml | 851 ng/ml | biomarker of hepatocyte injury |
Haptoglobin |
0.98 µg/ml | 14 µg/ml | biomarker |
MPO |
822 ng/ml | 25 ng/ml | myeloperoxidase |
*The least detectable dose was determined as the mean + 3 standard deviations of 20 blank readings.
Approximately 7% of all newly diagnosed cancers worldwide are liver cancers with the third most common cause of death worldwide (
In this study, we investigated the oncolytic efficiency of the vaccinia virus strain GLV-1h68 against the two hepatocellular carcinoma cell lines HuH7 and PLC in culture and the therapeutic efficacy in xenograft models. The results showed that GLV-1h68 was able to effectively infect, replicate in, and lyse hepatocellular carcinoma cells in culture. The efficiency of viral replication correlated well with degree of cell lysis and with expression of the marker
More importantly, the data clearly demonstrated that the optimal oncolytic effect of the GLV-1h68 is dependent on the interactions with the components of the tumor microenvironment, such as tumor vasculature and with the cells of the host immune system. The finding of the phenotypic switch in virus-infected PLC-xenograft tumors revealed a significant decrease in the number of blood vessels when compared to controls at day 10 after virus injection (P<0.01). The vascular density in infected HuH7 tumors, however, did not change in comparison to uninfected controls, which were non-hemorrhagic.
The protein profiling data of infected and uninfected PLC tumors (
Injection of GLV-1h68 to tumorous mice led to inhibition of tumor growth in both HuH7 and PLC xenografts at 21 dpi and beyond. Therefore, we investigated which components of the tumor microenvironment may play a crucial role in the optimal oncolytic effect of GLV-1h68-virus. In this context, we analyzed the virus colonization and the presence of host immune cells in the tumor tissues of virus-infected and uninfected HuH7- and PLC-tumor bearing mice at 7 and 10 dpi (
In summary, use of GLV-1h68 strain demonstrated outstanding anti-tumor and anti-vascular effects in PLC and lesser efficacy in HuH7 hepatocellular carcinoma xenografts. Therefore we propose that GLV-1h68 strain is a very potent live drug in preclinical studies to be used soon for the treatment of primary liver cancer in humans. Moreover, results of a Phase 1 study of intravenous administration of GL-ONC1 (GLV-1h68) Vaccinia virus in patients with advanced solid cancer demonstrated acceptable safety, preliminary evidence of anticancer activity and virus replication in several patients (positive for GL-ONC1 viral plaque assay and GFP imaging; (
Figure S1 shows the effects of GLV-1h68 virus infection on HuH7 and PLC cells. Hepatocellular carcinoma cells HuH7 (
(TIF)
We thank Ms. J. Langbein, Mr. Jason Aguilar, Mr. Terry Trevino, and Mrs. Irina Smirnow for excellent technical support and Dr. Z. Sokolovic for critical reading of the manuscript.